README / index.html
Hermes
auto-refresh: 2026-11-01 (188 models)
282f807
Raw History Blame Contribute Delete
147 kB
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>llm-bench.io — Local LLM Leaderboard</title>
<meta name="description" content="Live leaderboard of local LLMs benchmarked on real hardware — quality, speed, and VRAM requirements. Powered by llm-bench.io.">
<style>
:root{
--bg:#0b0e14; --panel:#11151f; --panel2:#161b28; --border:#232a3b;
--text:#e8ecf4; --muted:#8b94a7; --accent:#5b8cff; --accent2:#7c5bff;
--good:#4ade80; --mid:#facc15; --bad:#f87171;
}
*{box-sizing:border-box;margin:0;padding:0}
body{background:var(--bg);color:var(--text);font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,Helvetica,Arial,sans-serif;line-height:1.5;padding:24px 16px 60px}
.wrap{max-width:1180px;margin:0 auto}
header{display:flex;align-items:baseline;gap:12px;flex-wrap:wrap;margin-bottom:6px}
h1{font-size:24px;font-weight:700}
h1 .dot{color:var(--accent)}
.sub{color:var(--muted);font-size:13px;margin-bottom:18px}
.sub a{color:var(--accent);text-decoration:none}
.sub a:hover{text-decoration:underline}
.toolbar{display:flex;gap:10px;flex-wrap:wrap;align-items:center;margin-bottom:14px}
.chips{display:flex;gap:6px;flex-wrap:wrap}
.chip{background:var(--panel);border:1px solid var(--border);color:var(--muted);border-radius:999px;padding:5px 12px;font-size:12.5px;cursor:pointer;user-select:none;transition:all .15s}
.chip:hover{color:var(--text);border-color:#3a445c}
.chip.active{background:var(--accent);border-color:var(--accent);color:#fff}
.search{margin-left:auto;background:var(--panel);border:1px solid var(--border);border-radius:8px;color:var(--text);padding:7px 12px;font-size:13px;min-width:200px}
.search::placeholder{color:var(--muted)}
.meta{color:var(--muted);font-size:12px;margin-bottom:10px;display:flex;gap:16px;flex-wrap:wrap}
.meta b{color:var(--text);font-weight:600}
.card{background:var(--panel);border:1px solid var(--border);border-radius:12px;overflow:auto}
table{width:100%;border-collapse:collapse;font-size:13px;min-width:900px}
th,td{padding:9px 12px;text-align:left;border-bottom:1px solid var(--border);white-space:nowrap}
th{position:sticky;top:0;background:var(--panel2);color:var(--muted);font-size:11.5px;text-transform:uppercase;letter-spacing:.04em;cursor:pointer;user-select:none;z-index:1}
th:hover{color:var(--text)}
th .arrow{font-size:10px;margin-left:3px;opacity:.7}
td.num,th.num{text-align:right;font-variant-numeric:tabular-nums}
tr:hover td{background:rgba(91,140,255,.05)}
.model-cell{display:flex;flex-direction:column;gap:1px;max-width:340px}
.model-name{font-weight:600;color:var(--text);overflow:hidden;text-overflow:ellipsis}
.model-hw{color:var(--muted);font-size:11.5px;overflow:hidden;text-overflow:ellipsis}
.vram{color:var(--accent2);font-weight:600}
.qbar{display:inline-block;width:52px;height:5px;border-radius:3px;background:var(--border);vertical-align:middle;margin-right:7px;overflow:hidden}
.qbar i{display:block;height:100%;border-radius:3px}
.sc{color:var(--text)}
.sc.dim{color:var(--muted)}
.updated{color:var(--muted);font-size:11.5px}
footer{margin-top:22px;color:var(--muted);font-size:12px;display:flex;gap:8px;flex-wrap:wrap;align-items:center}
footer a{color:var(--accent);text-decoration:none}
footer a:hover{text-decoration:underline}
.empty{padding:40px;text-align:center;color:var(--muted)}
@media(max-width:700px){ .search{margin-left:0;width:100%} }
</style>
</head>
<body>
<div class="wrap">
<header>
<h1>llm-bench<span class="dot">.io</span> — Local LLM Leaderboard</h1>
</header>
<div class="sub">
Real-world benchmarks from actual users — LLM-judged <b>quality</b> and <b>speed</b> across coding, agent, role-play &amp; research tasks, grouped by VRAM.
<a href="https://llm-bench.io" target="_blank" rel="noopener">Run your own test →</a>
</div>
<div class="toolbar">
<div class="chips" id="vramChips"></div>
<input class="search" id="search" type="search" placeholder="Filter model name…" aria-label="Filter models">
</div>
<div class="meta" id="meta">
<span><b>188</b> models</span><span><b>1726</b> submissions in window</span><span>of <b>4076</b> total on llm-bench.io</span>
</div>
<div class="card">
<table>
<thead>
<tr>
<th data-sort="model">Model</th>
<th data-sort="vram" class="num">VRAM</th>
<th data-sort="runs" class="num">Runs</th>
<th data-sort="quality" class="num">Quality</th>
<th data-sort="speed" class="num">Speed (tok/s)</th>
<th data-sort="coding_agent" class="num">Coding</th>
<th data-sort="openclaw" class="num">Agent</th>
<th data-sort="roleplay" class="num">Role-Play</th>
<th data-sort="research" class="num">Research</th>
<th data-sort="updated" class="num">Updated</th>
</tr>
</thead>
<tbody id="rows">
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B" target="_blank" rel="noopener">Qwen3.8-27B</a>
<span class="model-hw">NVIDIA GeForce RTX 3090</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:89%;background:#4ade80"></i></span><b>88.6</b></td>
<td class="num">49.0</td>
<td class="num sc">84.2</td> <td class="num sc">89.6</td> <td class="num sc">92.4</td> <td class="num sc">88.3</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth-qwen3.8%3ABALANCED" target="_blank" rel="noopener">unsloth-qwen3.8:BALANCED</a>
<span class="model-hw">AMD Radeon Graphics</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:87%;background:#4ade80"></i></span><b>86.6</b></td>
<td class="num">38.1</td>
<td class="num sc">80.0</td> <td class="num sc">87.8</td> <td class="num sc">92.0</td> <td class="num sc">86.7</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-Turbo-Brilliance-50X-RI-GSQ-RCO-IQ3_XXS-mtp" target="_blank" rel="noopener">Qwen3.8-27B-Turbo-Brilliance-50X-RI-GSQ-RCO-IQ3_XXS-mtp</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:86%;background:#4ade80"></i></span><b>86.2</b></td>
<td class="num">37.5</td>
<td class="num sc">83.8</td> <td class="num sc">86.9</td> <td class="num sc">87.8</td> <td class="num sc">86.2</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ukisai%2FSwift-1.5-Qwen3.8-27B-GGUF%3ASwift-1.5-Qwen3.8-27B-Q4_K_S" target="_blank" rel="noopener">ukisai/Swift-1.5-Qwen3.8-27B-GGUF:Swift-1.5-Qwen3.8-27B-Q4_K_S</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:86%;background:#4ade80"></i></span><b>85.8</b></td>
<td class="num">72.4</td>
<td class="num sc">82.3</td> <td class="num sc">81.9</td> <td class="num sc">90.4</td> <td class="num sc">88.7</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-iq2_xs" target="_blank" rel="noopener">qwen3.8-flash-next-iq2_xs</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:86%;background:#4ade80"></i></span><b>85.7</b></td>
<td class="num">75.7</td>
<td class="num sc">80.7</td> <td class="num sc">87.3</td> <td class="num sc">89.4</td> <td class="num sc">85.4</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ornith-1.5-35B-Q4_K_M" target="_blank" rel="noopener">Ornith-1.5-35B-Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 5060</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:86%;background:#4ade80"></i></span><b>85.7</b></td>
<td class="num">52.2</td>
<td class="num sc">79.9</td> <td class="num sc">84.8</td> <td class="num sc">91.5</td> <td class="num sc">86.7</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS" target="_blank" rel="noopener">Tiel-Coder-35B-A3B-MTP-UD-IQ3_XXS</a>
<span class="model-hw">NVIDIA GeForce RTX 5060</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#4ade80"></i></span><b>85.3</b></td>
<td class="num">77.0</td>
<td class="num sc">79.9</td> <td class="num sc">86.5</td> <td class="num sc">88.2</td> <td class="num sc">86.5</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=bonsai-2%3A27b" target="_blank" rel="noopener">bonsai-2:27b</a>
<span class="model-hw">AMD Radeon RX 9060 XT + AMD Radeon RX 7600/7600 XT/7600M XT/7600S/7700S / PRO W7600</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#4ade80"></i></span><b>85.3</b></td>
<td class="num">25.4</td>
<td class="num sc">77.2</td> <td class="num sc">88.2</td> <td class="num sc">88.8</td> <td class="num sc">87.1</td>
<td class="num updated">2026-10-25</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-27b-ud-q3_k_xl" target="_blank" rel="noopener">qwen3.8-27b-ud-q3_k_xl</a>
<span class="model-hw">NVIDIA GeForce RTX 5080</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#4ade80"></i></span><b>85.1</b></td>
<td class="num">52.3</td>
<td class="num sc">77.6</td> <td class="num sc">82.2</td> <td class="num sc">92.0</td> <td class="num sc">88.7</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=peculiar-ragdoll%2FCyber-Tiel-Coder-35B-A3B-GGUF-MTP" target="_blank" rel="noopener">peculiar-ragdoll/Cyber-Tiel-Coder-35B-A3B-GGUF-MTP</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#facc15"></i></span><b>84.9</b></td>
<td class="num">74.0</td>
<td class="num sc">75.9</td> <td class="num sc">88.7</td> <td class="num sc">85.9</td> <td class="num sc">89.0</td>
<td class="num updated">2026-10-24</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Swift-1.5-Qwen3.8-27B-GSQ-RCO-IQ3_XXS-mtp" target="_blank" rel="noopener">Swift-1.5-Qwen3.8-27B-GSQ-RCO-IQ3_XXS-mtp</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">15</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#facc15"></i></span><b>84.8</b></td>
<td class="num">48.4</td>
<td class="num sc">75.1</td> <td class="num sc">85.0</td> <td class="num sc">93.3</td> <td class="num sc">85.7</td>
<td class="num updated">2026-10-18</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp-ASCII-Condensed" target="_blank" rel="noopener">Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp-ASCII-Condensed</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#facc15"></i></span><b>84.8</b></td>
<td class="num">44.6</td>
<td class="num sc">74.4</td> <td class="num sc">86.4</td> <td class="num sc">90.7</td> <td class="num sc">87.7</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-oQ6e-mtp" target="_blank" rel="noopener">Qwen3.8-27B-oQ6e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#facc15"></i></span><b>84.7</b></td>
<td class="num">50.7</td>
<td class="num sc">84.3</td> <td class="num sc">86.0</td> <td class="num sc">84.8</td> <td class="num sc">83.8</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=mlx-community%2FQwen3.8-27B-4bit" target="_blank" rel="noopener">mlx-community/Qwen3.8-27B-4bit</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:85%;background:#facc15"></i></span><b>84.6</b></td>
<td class="num">82.9</td>
<td class="num sc">74.2</td> <td class="num sc">84.6</td> <td class="num sc">91.5</td> <td class="num sc">88.2</td>
<td class="num updated">2026-10-21</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ornith-1.5-35B-A3B-oQ8e-mtp" target="_blank" rel="noopener">Ornith-1.5-35B-A3B-oQ8e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.5</b></td>
<td class="num">126.3</td>
<td class="num sc">81.7</td> <td class="num sc">87.2</td> <td class="num sc">86.4</td> <td class="num sc">82.8</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ukisai%2FSwift-Bonsai-2-GGUF%3ATQ1_0" target="_blank" rel="noopener">ukisai/Swift-Bonsai-2-GGUF:TQ1_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.5</b></td>
<td class="num">64.7</td>
<td class="num sc">79.7</td> <td class="num sc">86.1</td> <td class="num sc">89.4</td> <td class="num sc">82.7</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF-RVN-Q6_K-multilingual-mtp" target="_blank" rel="noopener">Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF-RVN-Q6_K-multilingual-mtp</a>
<span class="model-hw">NVIDIA RTX A6000</span></div></td>
<td class="num vram">47 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.5</b></td>
<td class="num">27.3</td>
<td class="num sc">79.7</td> <td class="num sc">84.1</td> <td class="num sc">88.5</td> <td class="num sc">85.7</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-oQ4e-mtp" target="_blank" rel="noopener">Qwen3.8-27B-oQ4e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">8</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.2</b></td>
<td class="num">41.8</td>
<td class="num sc">75.8</td> <td class="num sc">84.2</td> <td class="num sc">91.2</td> <td class="num sc">85.7</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-GSQ-RCO-IQ3_S-recovered" target="_blank" rel="noopener">Qwen3.8-27B-GSQ-RCO-IQ3_S-recovered</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.2</b></td>
<td class="num">35.5</td>
<td class="num sc">77.8</td> <td class="num sc">79.8</td> <td class="num sc">91.9</td> <td class="num sc">87.3</td>
<td class="num updated">2026-10-06</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF-RVN-Q5_K_M-mtp" target="_blank" rel="noopener">Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF-RVN-Q5_K_M-mtp</a>
<span class="model-hw">NVIDIA RTX A6000</span></div></td>
<td class="num vram">47 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.2</b></td>
<td class="num">29.7</td>
<td class="num sc">78.5</td> <td class="num sc">83.9</td> <td class="num sc">88.4</td> <td class="num sc">86.0</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-27b-gsq3" target="_blank" rel="noopener">qwen3.8-27b-gsq3</a>
<span class="model-hw">NVIDIA GeForce RTX 5060</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.1</b></td>
<td class="num">114.4</td>
<td class="num sc">77.9</td> <td class="num sc">84.8</td> <td class="num sc">87.2</td> <td class="num sc">86.5</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=huihui-swift15-qwen3.8-27b" target="_blank" rel="noopener">huihui-swift15-qwen3.8-27b</a>
<span class="model-hw">NVIDIA GeForce RTX 3080 + NVIDIA GeForce RTX 5080</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.0</b></td>
<td class="num">118.8</td>
<td class="num sc">78.7</td> <td class="num sc">83.4</td> <td class="num sc">88.9</td> <td class="num sc">84.8</td>
<td class="num updated">2026-10-26</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=swift-1.5-iq3_xxs" target="_blank" rel="noopener">swift-1.5-iq3_xxs</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>84.0</b></td>
<td class="num">53.1</td>
<td class="num sc">78.1</td> <td class="num sc">84.8</td> <td class="num sc">88.5</td> <td class="num sc">84.7</td>
<td class="num updated">2026-10-03</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-Flash-Next-oQ5e-mtp" target="_blank" rel="noopener">Qwen3.8-Flash-Next-oQ5e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>83.9</b></td>
<td class="num">70.2</td>
<td class="num sc">73.9</td> <td class="num sc">85.1</td> <td class="num sc">92.1</td> <td class="num sc">84.7</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-MXFP4" target="_blank" rel="noopener">Qwen3.8-MXFP4</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>83.9</b></td>
<td class="num">66.4</td>
<td class="num sc">74.9</td> <td class="num sc">85.6</td> <td class="num sc">89.1</td> <td class="num sc">86.0</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8%3A27b-q4_K_M" target="_blank" rel="noopener">qwen3.8:27b-q4_K_M</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>83.9</b></td>
<td class="num">26.4</td>
<td class="num sc">78.7</td> <td class="num sc">80.5</td> <td class="num sc">92.3</td> <td class="num sc">84.2</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FMuse-Glimmer-30B-GGUF%3AUD-IQ3_XXS" target="_blank" rel="noopener">unsloth/Muse-Glimmer-30B-GGUF:UD-IQ3_XXS</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>83.7</b></td>
<td class="num">37.6</td>
<td class="num sc">69.0</td> <td class="num sc">90.6</td> <td class="num sc">91.6</td> <td class="num sc">83.4</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=swift-1.5-qwen3.8-27b-uncensored" target="_blank" rel="noopener">swift-1.5-qwen3.8-27b-uncensored</a>
<span class="model-hw">NVIDIA GeForce RTX 5060 Ti + NVIDIA GeForce RTX 5090</span></div></td>
<td class="num vram">31 GB</td>
<td class="num">19</td>
<td class="num"><span class="qbar"><i style="width:84%;background:#facc15"></i></span><b>83.6</b></td>
<td class="num">284.1</td>
<td class="num sc">76.7</td> <td class="num sc">82.3</td> <td class="num sc">91.0</td> <td class="num sc">84.3</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Swift-1.5-Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp" target="_blank" rel="noopener">Swift-1.5-Qwen3.8-27B-GSQ-RCO-IQ3_S-mtp</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.4</b></td>
<td class="num">39.6</td>
<td class="num sc">80.6</td> <td class="num sc">83.5</td> <td class="num sc">83.7</td> <td class="num sc">86.0</td>
<td class="num updated">2026-10-18</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-27B-gguf%3APTQ1_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-27B-gguf:PTQ1_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.3</b></td>
<td class="num">58.9</td>
<td class="num sc">74.6</td> <td class="num sc">87.2</td> <td class="num sc">87.4</td> <td class="num sc">84.1</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ukisai%2FSwift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF%3AIQ3_S-mtp" target="_blank" rel="noopener">ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_S-mtp</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.3</b></td>
<td class="num">46.9</td>
<td class="num sc">73.3</td> <td class="num sc">83.1</td> <td class="num sc">89.7</td> <td class="num sc">87.2</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FMuse-Glimmer-30B-GGUF%3AUD-Q4_K_XL" target="_blank" rel="noopener">unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.3</b></td>
<td class="num">34.4</td>
<td class="num sc">75.4</td> <td class="num sc">87.1</td> <td class="num sc">86.2</td> <td class="num sc">84.4</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B.APEX-I-MiniPlus-V2.1" target="_blank" rel="noopener">Tiel-Coder-35B-A3B.APEX-I-MiniPlus-V2.1</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">12</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.2</b></td>
<td class="num">78.9</td>
<td class="num sc">77.7</td> <td class="num sc">88.1</td> <td class="num sc">82.8</td> <td class="num sc">84.4</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ornith-1.5-35B-A3B-oQ4e-mtp" target="_blank" rel="noopener">Ornith-1.5-35B-A3B-oQ4e-mtp</a>
<span class="model-hw">Apple M5 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.2</b></td>
<td class="num">74.2</td>
<td class="num sc">77.7</td> <td class="num sc">85.2</td> <td class="num sc">85.0</td> <td class="num sc">84.9</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-oQ4e-mtp" target="_blank" rel="noopener">Qwen3.8-27B-oQ4e-mtp</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>83.0</b></td>
<td class="num">29.6</td>
<td class="num sc">76.8</td> <td class="num sc">80.7</td> <td class="num sc">89.5</td> <td class="num sc">84.9</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Nex-N2.5-mini-IQ3_XXS" target="_blank" rel="noopener">Nex-N2.5-mini-IQ3_XXS</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.9</b></td>
<td class="num">106.2</td>
<td class="num sc">77.4</td> <td class="num sc">87.5</td> <td class="num sc">85.3</td> <td class="num sc">81.5</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-27b-splash" target="_blank" rel="noopener">qwen3.8-27b-splash</a>
<span class="model-hw">Apple M4 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">97</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.7</b></td>
<td class="num">73.5</td>
<td class="num sc">77.3</td> <td class="num sc">78.2</td> <td class="num sc">90.0</td> <td class="num sc">85.3</td>
<td class="num updated">2026-10-03</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=nex-n2.5-mini" target="_blank" rel="noopener">nex-n2.5-mini</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.7</b></td>
<td class="num">66.1</td>
<td class="num sc">77.0</td> <td class="num sc">81.5</td> <td class="num sc">87.8</td> <td class="num sc">84.5</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-oQ8e-mtp" target="_blank" rel="noopener">Qwen3.8-27B-oQ8e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">32</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.7</b></td>
<td class="num">46.5</td>
<td class="num sc">74.3</td> <td class="num sc">84.2</td> <td class="num sc">87.7</td> <td class="num sc">84.6</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=IsValorum%2FOrnith-1.5-35B-A3B-APEX-I-NanoPlus-GGUF" target="_blank" rel="noopener">IsValorum/Ornith-1.5-35B-A3B-APEX-I-NanoPlus-GGUF</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.6</b></td>
<td class="num">104.3</td>
<td class="num sc">76.5</td> <td class="num sc">83.7</td> <td class="num sc">86.8</td> <td class="num sc">83.5</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ornith-1.5%3A35B" target="_blank" rel="noopener">ornith-1.5:35B</a>
<span class="model-hw">AMD Radeon Graphics</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:83%;background:#facc15"></i></span><b>82.6</b></td>
<td class="num">79.5</td>
<td class="num sc">73.9</td> <td class="num sc">87.8</td> <td class="num sc">82.9</td> <td class="num sc">86.0</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-unsloth-ud-q4_k_xl" target="_blank" rel="noopener">qwen3.8-flash-next-unsloth-ud-q4_k_xl</a>
<span class="model-hw">NVIDIA GeForce RTX 3090</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>82.3</b></td>
<td class="num">24.5</td>
<td class="num sc">63.9</td> <td class="num sc">86.5</td> <td class="num sc">92.5</td> <td class="num sc">86.4</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-iq3_xxs" target="_blank" rel="noopener">qwen3.8-flash-next-iq3_xxs</a>
<span class="model-hw">AMD Radeon RX 6950 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>82.2</b></td>
<td class="num">50.5</td>
<td class="num sc">66.0</td> <td class="num sc">86.0</td> <td class="num sc">91.5</td> <td class="num sc">85.3</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-abliterated-iq3_s" target="_blank" rel="noopener">qwen3.8-flash-next-abliterated-iq3_s</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>82.1</b></td>
<td class="num">57.5</td>
<td class="num sc">69.2</td> <td class="num sc">83.0</td> <td class="num sc">91.0</td> <td class="num sc">85.4</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ornith-1.5-35B-A3B.APEX-I-NanoPlus-ASCII-Condensed" target="_blank" rel="noopener">Ornith-1.5-35B-A3B.APEX-I-NanoPlus-ASCII-Condensed</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>82.0</b></td>
<td class="num">110.7</td>
<td class="num sc">78.9</td> <td class="num sc">76.2</td> <td class="num sc">85.3</td> <td class="num sc">87.8</td>
<td class="num updated">2026-10-09</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=muse-glimmer%3A30b-q4_K_M" target="_blank" rel="noopener">muse-glimmer:30b-q4_K_M</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>82.0</b></td>
<td class="num">27.5</td>
<td class="num sc">69.0</td> <td class="num sc">90.7</td> <td class="num sc">86.5</td> <td class="num sc">81.8</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-Flash-Next-oQ4e-mix6-mtp" target="_blank" rel="noopener">Qwen3.8-Flash-Next-oQ4e-mix6-mtp</a>
<span class="model-hw">Apple M4 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.9</b></td>
<td class="num">72.9</td>
<td class="num sc">74.0</td> <td class="num sc">82.1</td> <td class="num sc">87.4</td> <td class="num sc">84.2</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=SC117-Swift-IQ3_S" target="_blank" rel="noopener">SC117-Swift-IQ3_S</a>
<span class="model-hw">NVIDIA RTX A6000</span></div></td>
<td class="num vram">47 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.6</b></td>
<td class="num">94.9</td>
<td class="num sc">55.9</td> <td class="num sc">90.5</td> <td class="num sc">91.4</td> <td class="num sc">88.5</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-27B-gguf%3ATQ1_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-27B-gguf:TQ1_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.6</b></td>
<td class="num">57.0</td>
<td class="num sc">73.8</td> <td class="num sc">87.5</td> <td class="num sc">81.7</td> <td class="num sc">83.4</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ternary-Bonsai-2-27B-PTQ1_0" target="_blank" rel="noopener">Ternary-Bonsai-2-27B-PTQ1_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.6</b></td>
<td class="num">50.7</td>
<td class="num sc">62.0</td> <td class="num sc">87.4</td> <td class="num sc">91.1</td> <td class="num sc">85.9</td>
<td class="num updated">2026-10-08</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=muse-glimmer-30b" target="_blank" rel="noopener">muse-glimmer-30b</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.6</b></td>
<td class="num">30.8</td>
<td class="num sc">72.5</td> <td class="num sc">82.8</td> <td class="num sc">88.5</td> <td class="num sc">82.8</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=underdog-saluki-27b-1.0" target="_blank" rel="noopener">underdog-saluki-27b-1.0</a>
<span class="model-hw">AMD Radeon RX 6800/6800 XT / 6900 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:82%;background:#facc15"></i></span><b>81.5</b></td>
<td class="num">31.0</td>
<td class="num sc">70.5</td> <td class="num sc">83.8</td> <td class="num sc">87.9</td> <td class="num sc">83.8</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-27b-splash" target="_blank" rel="noopener">qwen3.8-27b-splash</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:81%;background:#facc15"></i></span><b>81.2</b></td>
<td class="num">78.6</td>
<td class="num sc">70.7</td> <td class="num sc">80.8</td> <td class="num sc">89.8</td> <td class="num sc">83.4</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ISTA-DASLab%2FQwen3.8-27B-GSQ-RCO-GGUF%3AIQ3_XXS-mtp" target="_blank" rel="noopener">ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF:IQ3_XXS-mtp</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:81%;background:#facc15"></i></span><b>81.2</b></td>
<td class="num">42.4</td>
<td class="num sc">64.6</td> <td class="num sc">85.7</td> <td class="num sc">88.8</td> <td class="num sc">85.7</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B-MLX-oQ6e-MTP" target="_blank" rel="noopener">Tiel-Coder-35B-A3B-MLX-oQ6e-MTP</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">8</td>
<td class="num"><span class="qbar"><i style="width:81%;background:#facc15"></i></span><b>80.7</b></td>
<td class="num">134.6</td>
<td class="num sc">74.6</td> <td class="num sc">82.4</td> <td class="num sc">81.7</td> <td class="num sc">84.1</td>
<td class="num updated">2026-10-17</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B-MLX-oQ4e-MTP" target="_blank" rel="noopener">Tiel-Coder-35B-A3B-MLX-oQ4e-MTP</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:80%;background:#facc15"></i></span><b>80.2</b></td>
<td class="num">174.6</td>
<td class="num sc">79.7</td> <td class="num sc">81.5</td> <td class="num sc">76.0</td> <td class="num sc">83.4</td>
<td class="num updated">2026-10-17</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-huihui-ud-q4_k_xl" target="_blank" rel="noopener">qwen3.8-flash-next-huihui-ud-q4_k_xl</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:80%;background:#facc15"></i></span><b>80.0</b></td>
<td class="num">33.0</td>
<td class="num sc">60.2</td> <td class="num sc">88.5</td> <td class="num sc">86.5</td> <td class="num sc">84.7</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FQwen3.6-35B-A3B-MTP-GGUF%3AUD-Q4_K_XL" target="_blank" rel="noopener">unsloth/Qwen3.6-35B-A3B-MTP-GGUF:UD-Q4_K_XL</a>
<span class="model-hw">Advanced Micro Devices, Inc. [AMD/ATI] Device 7551</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:80%;background:#facc15"></i></span><b>79.5</b></td>
<td class="num">101.0</td>
<td class="num sc">69.8</td> <td class="num sc">78.0</td> <td class="num sc">88.4</td> <td class="num sc">81.8</td>
<td class="num updated">2026-10-04</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.6-35B-A3B-GSQ-hybrid" target="_blank" rel="noopener">Qwen3.6-35B-A3B-GSQ-hybrid</a>
<span class="model-hw">NVIDIA GeForce RTX 5060</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:80%;background:#facc15"></i></span><b>79.5</b></td>
<td class="num">82.9</td>
<td class="num sc">69.8</td> <td class="num sc">81.0</td> <td class="num sc">88.5</td> <td class="num sc">78.6</td>
<td class="num updated">2026-10-04</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FQwen3.8-27B-GGUF%3AUD-Q4_K_M" target="_blank" rel="noopener">unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_M</a>
<span class="model-hw">Apple M5 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:79%;background:#facc15"></i></span><b>79.3</b></td>
<td class="num">36.1</td>
<td class="num sc">71.0</td> <td class="num sc">77.5</td> <td class="num sc">87.1</td> <td class="num sc">81.7</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-unsloth-ud-iq4_xs" target="_blank" rel="noopener">qwen3.8-flash-next-unsloth-ud-iq4_xs</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:79%;background:#facc15"></i></span><b>78.7</b></td>
<td class="num">46.9</td>
<td class="num sc">51.8</td> <td class="num sc">87.1</td> <td class="num sc">90.4</td> <td class="num sc">85.2</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=swift-1.5-iq2_xs" target="_blank" rel="noopener">swift-1.5-iq2_xs</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:78%;background:#facc15"></i></span><b>77.7</b></td>
<td class="num">80.2</td>
<td class="num sc">61.9</td> <td class="num sc">72.9</td> <td class="num sc">92.3</td> <td class="num sc">83.8</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.6-35B-A3B-UD-IQ2_XXS" target="_blank" rel="noopener">Qwen3.6-35B-A3B-UD-IQ2_XXS</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">12</td>
<td class="num"><span class="qbar"><i style="width:78%;background:#facc15"></i></span><b>77.6</b></td>
<td class="num">209.8</td>
<td class="num sc">71.9</td> <td class="num sc">75.6</td> <td class="num sc">83.3</td> <td class="num sc">79.7</td>
<td class="num updated">2026-10-02</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FQwen3.6-35B-A3B-MTP-GGUF" target="_blank" rel="noopener">unsloth/Qwen3.6-35B-A3B-MTP-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:78%;background:#facc15"></i></span><b>77.6</b></td>
<td class="num">52.3</td>
<td class="num sc">75.8</td> <td class="num sc">74.5</td> <td class="num sc">83.1</td> <td class="num sc">77.1</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=pfeifferj%2FOrnith-1.5-35B-A3B-GSQ-RCO-GGUF%3A3bit" target="_blank" rel="noopener">pfeifferj/Ornith-1.5-35B-A3B-GSQ-RCO-GGUF:3bit</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:77%;background:#facc15"></i></span><b>77.4</b></td>
<td class="num">88.5</td>
<td class="num sc">73.9</td> <td class="num sc">63.7</td> <td class="num sc">86.9</td> <td class="num sc">85.1</td>
<td class="num updated">2026-10-08</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FQwen3.8-27B-GGUF%3AUD-Q4_K_XL" target="_blank" rel="noopener">unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL</a>
<span class="model-hw">Advanced Micro Devices, Inc. [AMD/ATI] Device 7551</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:77%;background:#facc15"></i></span><b>77.2</b></td>
<td class="num">35.8</td>
<td class="num sc">44.8</td> <td class="num sc">87.8</td> <td class="num sc">89.7</td> <td class="num sc">86.3</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-MTPLX-Optimized-Speed" target="_blank" rel="noopener">Qwen3.8-27B-MTPLX-Optimized-Speed</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:77%;background:#facc15"></i></span><b>77.0</b></td>
<td class="num">24.1</td>
<td class="num sc">63.1</td> <td class="num sc">82.7</td> <td class="num sc">80.7</td> <td class="num sc">81.6</td>
<td class="num updated">2026-10-18</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B.APEX-I-MiniPlus-V2.1" target="_blank" rel="noopener">Tiel-Coder-35B-A3B.APEX-I-MiniPlus-V2.1</a>
<span class="model-hw">NVIDIA GeForce RTX 4050 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:77%;background:#facc15"></i></span><b>76.7</b></td>
<td class="num">26.4</td>
<td class="num sc">68.4</td> <td class="num sc">78.4</td> <td class="num sc">84.3</td> <td class="num sc">75.8</td>
<td class="num updated">2026-10-25</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth-qwen3.8%3ATOOLING" target="_blank" rel="noopener">unsloth-qwen3.8:TOOLING</a>
<span class="model-hw">AMD Radeon Graphics</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:77%;background:#facc15"></i></span><b>76.6</b></td>
<td class="num">34.1</td>
<td class="num sc">59.2</td> <td class="num sc">81.6</td> <td class="num sc">83.9</td> <td class="num sc">81.5</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=tiel-coder-35b" target="_blank" rel="noopener">tiel-coder-35b</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>76.2</b></td>
<td class="num">76.0</td>
<td class="num sc">75.4</td> <td class="num sc">79.4</td> <td class="num sc">73.2</td> <td class="num sc">76.7</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=MiniCPM-V-4.7-35B-A3B.APEX-I-NanoPlus-V2.1" target="_blank" rel="noopener">MiniCPM-V-4.7-35B-A3B.APEX-I-NanoPlus-V2.1</a>
<span class="model-hw">NVIDIA GeForce RTX 5060</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>76.1</b></td>
<td class="num">92.7</td>
<td class="num sc">61.8</td> <td class="num sc">84.6</td> <td class="num sc">79.9</td> <td class="num sc">78.0</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=swift-1.5-iq3_s" target="_blank" rel="noopener">swift-1.5-iq3_s</a>
<span class="model-hw">NVIDIA GeForce RTX 5090</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>76.0</b></td>
<td class="num">206.7</td>
<td class="num sc">51.7</td> <td class="num sc">83.4</td> <td class="num sc">86.8</td> <td class="num sc">82.1</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gpt-oss%3A20b" target="_blank" rel="noopener">gpt-oss:20b</a>
<span class="model-hw">AMD Radeon RX 9060 XT + AMD Radeon RX 7600/7600 XT/7600M XT/7600S/7700S / PRO W7600</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>76.0</b></td>
<td class="num">64.5</td>
<td class="num sc">73.0</td> <td class="num sc">83.0</td> <td class="num sc">72.8</td> <td class="num sc">75.3</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-27B-gguf%3APQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-27B-gguf:PQ2_0</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.8</b></td>
<td class="num">41.5</td>
<td class="num sc">48.2</td> <td class="num sc">84.4</td> <td class="num sc">86.0</td> <td class="num sc">84.5</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=nvidia-nemotron-3.5-lightning-30b-a3b" target="_blank" rel="noopener">nvidia-nemotron-3.5-lightning-30b-a3b</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.7</b></td>
<td class="num">111.4</td>
<td class="num sc">66.5</td> <td class="num sc">85.9</td> <td class="num sc">71.5</td> <td class="num sc">78.7</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.6-35B-A3B-Uncensored-Heretic-MLX-4bit" target="_blank" rel="noopener">Qwen3.6-35B-A3B-Uncensored-Heretic-MLX-4bit</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.6</b></td>
<td class="num">121.2</td>
<td class="num sc">71.4</td> <td class="num sc">73.6</td> <td class="num sc">83.6</td> <td class="num sc">73.8</td>
<td class="num updated">2026-10-21</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ornith15-heretic-mini" target="_blank" rel="noopener">ornith15-heretic-mini</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.6</b></td>
<td class="num">72.8</td>
<td class="num sc">59.3</td> <td class="num sc">81.3</td> <td class="num sc">86.5</td> <td class="num sc">75.4</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-27b-gsq-rco%40iq2_xs" target="_blank" rel="noopener">qwen3.8-27b-gsq-rco@iq2_xs</a>
<span class="model-hw">AMD Radeon RX 7900 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.6</b></td>
<td class="num">31.2</td>
<td class="num sc">78.0</td> <td class="num sc">64.6</td> <td class="num sc">82.7</td> <td class="num sc">77.2</td>
<td class="num updated">2026-10-08</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen38-27b-turbo" target="_blank" rel="noopener">qwen38-27b-turbo</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.6</b></td>
<td class="num">29.5</td>
<td class="num sc">71.2</td> <td class="num sc">75.6</td> <td class="num sc">78.9</td> <td class="num sc">76.7</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwopus3.8-27B-Flash-V2-oQ8e-mtp" target="_blank" rel="noopener">Qwopus3.8-27B-Flash-V2-oQ8e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:76%;background:#facc15"></i></span><b>75.5</b></td>
<td class="num">46.5</td>
<td class="num sc">56.5</td> <td class="num sc">80.1</td> <td class="num sc">84.9</td> <td class="num sc">80.6</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=nemotron-3.5-bolt" target="_blank" rel="noopener">nemotron-3.5-bolt</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:75%;background:#facc15"></i></span><b>75.3</b></td>
<td class="num">112.4</td>
<td class="num sc">60.2</td> <td class="num sc">83.7</td> <td class="num sc">77.4</td> <td class="num sc">80.0</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.6-35B-A3B-4bit" target="_blank" rel="noopener">Qwen3.6-35B-A3B-4bit</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:75%;background:#facc15"></i></span><b>75.1</b></td>
<td class="num">118.4</td>
<td class="num sc">69.3</td> <td class="num sc">77.4</td> <td class="num sc">75.0</td> <td class="num sc">78.7</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=IsValorum%2FOccamy-1.0-APEX-I-NanoPlus-GGUF" target="_blank" rel="noopener">IsValorum/Occamy-1.0-APEX-I-NanoPlus-GGUF</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">8</td>
<td class="num"><span class="qbar"><i style="width:75%;background:#f87171"></i></span><b>74.9</b></td>
<td class="num">104.0</td>
<td class="num sc">56.0</td> <td class="num sc">75.0</td> <td class="num sc">84.5</td> <td class="num sc">84.0</td>
<td class="num updated">2026-10-06</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-Uncensored-MLX" target="_blank" rel="noopener">Qwen3.8-27B-Uncensored-MLX</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:75%;background:#f87171"></i></span><b>74.7</b></td>
<td class="num">26.2</td>
<td class="num sc">54.3</td> <td class="num sc">77.8</td> <td class="num sc">83.8</td> <td class="num sc">83.0</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-unsloth-ud-iq4_xs" target="_blank" rel="noopener">qwen3.8-flash-next-unsloth-ud-iq4_xs</a>
<span class="model-hw">NVIDIA GeForce RTX 3090</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:75%;background:#f87171"></i></span><b>74.6</b></td>
<td class="num">46.3</td>
<td class="num sc">38.0</td> <td class="num sc">90.8</td> <td class="num sc">94.2</td> <td class="num sc">75.5</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ornith-1.5-35B-A3B.APEX-I-NanoPlus" target="_blank" rel="noopener">Ornith-1.5-35B-A3B.APEX-I-NanoPlus</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.5</b></td>
<td class="num">110.5</td>
<td class="num sc">81.1</td> <td class="num sc">61.2</td> <td class="num sc">69.2</td> <td class="num sc">86.4</td>
<td class="num updated">2026-10-18</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=protoLabsAI%2FOrnith-1.5-9B-MTP-GGUF%3AQ5_K_M" target="_blank" rel="noopener">protoLabsAI/Ornith-1.5-9B-MTP-GGUF:Q5_K_M</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.5</b></td>
<td class="num">78.9</td>
<td class="num sc">53.8</td> <td class="num sc">76.4</td> <td class="num sc">82.5</td> <td class="num sc">85.5</td>
<td class="num updated">2026-10-06</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-flash-next-gyro-s-abliterated" target="_blank" rel="noopener">qwen3.8-flash-next-gyro-s-abliterated</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.4</b></td>
<td class="num">77.3</td>
<td class="num sc">60.5</td> <td class="num sc">71.8</td> <td class="num sc">87.3</td> <td class="num sc">78.2</td>
<td class="num updated">2026-10-23</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=recursivesi%2Fflint" target="_blank" rel="noopener">recursivesi/flint</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.3</b></td>
<td class="num">107.4</td>
<td class="num sc">54.4</td> <td class="num sc">78.7</td> <td class="num sc">84.4</td> <td class="num sc">79.8</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwopus3.8-27B-Flash-V2-oQ4e-mtp" target="_blank" rel="noopener">Qwopus3.8-27B-Flash-V2-oQ4e-mtp</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.3</b></td>
<td class="num">66.9</td>
<td class="num sc">56.9</td> <td class="num sc">76.8</td> <td class="num sc">84.3</td> <td class="num sc">79.3</td>
<td class="num updated">2026-10-11</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-UD-Q3_K_XL" target="_blank" rel="noopener">Qwen3.8-27B-UD-Q3_K_XL</a>
<span class="model-hw">NVIDIA GeForce RTX 5070 Ti</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">12</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>74.2</b></td>
<td class="num">60.5</td>
<td class="num sc">42.1</td> <td class="num sc">85.2</td> <td class="num sc">90.2</td> <td class="num sc">79.2</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=openbmb%2FMiniCPM-V-4.7-35B-A3B-gguf" target="_blank" rel="noopener">openbmb/MiniCPM-V-4.7-35B-A3B-gguf</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:74%;background:#f87171"></i></span><b>73.7</b></td>
<td class="num">51.8</td>
<td class="num sc">66.7</td> <td class="num sc">83.2</td> <td class="num sc">73.3</td> <td class="num sc">71.7</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=text-model" target="_blank" rel="noopener">text-model</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA RTX 2000 Ada Generation</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">8</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>73.2</b></td>
<td class="num">91.3</td>
<td class="num sc">68.0</td> <td class="num sc">72.0</td> <td class="num sc">81.2</td> <td class="num sc">71.7</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=cyankiwi%2Fgemma-4-31B-it-AWQ-4bit" target="_blank" rel="noopener">cyankiwi/gemma-4-31B-it-AWQ-4bit</a>
<span class="model-hw">NVIDIA GeForce RTX 3080 Ti + NVIDIA GeForce RTX 3080</span></div></td>
<td class="num vram">20 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>73.2</b></td>
<td class="num">32.3</td>
<td class="num sc">61.5</td> <td class="num sc">82.5</td> <td class="num sc">79.1</td> <td class="num sc">69.8</td>
<td class="num updated">2026-10-22</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Cyber-Tiel-Coder-35B-A3B-UD-IQ3_S" target="_blank" rel="noopener">Cyber-Tiel-Coder-35B-A3B-UD-IQ3_S</a>
<span class="model-hw">AMD Radeon 680M</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>72.9</b></td>
<td class="num">11.2</td>
<td class="num sc">55.9</td> <td class="num sc">80.2</td> <td class="num sc">81.5</td> <td class="num sc">74.0</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NeoHorse-1-4B-GGUF" target="_blank" rel="noopener">NeoHorse-1-4B-GGUF</a>
<span class="model-hw">NVIDIA RTX A6000</span></div></td>
<td class="num vram">47 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>72.8</b></td>
<td class="num">104.9</td>
<td class="num sc">57.2</td> <td class="num sc">81.2</td> <td class="num sc">74.9</td> <td class="num sc">77.9</td>
<td class="num updated">2026-10-22</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K-DFLASH-NVFP4" target="_blank" rel="noopener">NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K-DFLASH-NVFP4</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>72.7</b></td>
<td class="num">65.6</td>
<td class="num sc">65.5</td> <td class="num sc">81.5</td> <td class="num sc">80.4</td> <td class="num sc">63.6</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-9B-Q4_K_M" target="_blank" rel="noopener">Qwen3.8-9B-Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:73%;background:#f87171"></i></span><b>72.7</b></td>
<td class="num">34.5</td>
<td class="num sc">57.8</td> <td class="num sc">81.3</td> <td class="num sc">78.8</td> <td class="num sc">72.9</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Huihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF" target="_blank" rel="noopener">Huihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>72.5</b></td>
<td class="num">57.3</td>
<td class="num sc">72.8</td> <td class="num sc">57.1</td> <td class="num sc">85.2</td> <td class="num sc">74.9</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ornith-1.5-9b" target="_blank" rel="noopener">ornith-1.5-9b</a>
<span class="model-hw">AMD Radeon RX 9050 / 9060 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">12</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>72.2</b></td>
<td class="num">92.1</td>
<td class="num sc">60.3</td> <td class="num sc">62.6</td> <td class="num sc">83.7</td> <td class="num sc">82.4</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K-DFLASH-BF16" target="_blank" rel="noopener">NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K-DFLASH-BF16</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>72.2</b></td>
<td class="num">61.2</td>
<td class="num sc">65.9</td> <td class="num sc">77.2</td> <td class="num sc">80.5</td> <td class="num sc">65.1</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.6-35B-A3B-oQ4e-mtp" target="_blank" rel="noopener">Qwen3.6-35B-A3B-oQ4e-mtp</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">21</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>72.0</b></td>
<td class="num">97.8</td>
<td class="num sc">66.9</td> <td class="num sc">65.4</td> <td class="num sc">79.7</td> <td class="num sc">76.2</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Tiel-Coder-35B-A3B-MLX-oQ4e-MTP" target="_blank" rel="noopener">Tiel-Coder-35B-A3B-MLX-oQ4e-MTP</a>
<span class="model-hw">Apple M4 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>71.8</b></td>
<td class="num">153.1</td>
<td class="num sc">67.6</td> <td class="num sc">83.7</td> <td class="num sc">53.4</td> <td class="num sc">82.4</td>
<td class="num updated">2026-10-22</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=mistralai%2FMinistral-3-8B-Instruct-2512-GGUF%3AQ4_K_M" target="_blank" rel="noopener">mistralai/Ministral-3-8B-Instruct-2512-GGUF:Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>71.6</b></td>
<td class="num">62.8</td>
<td class="num sc">56.5</td> <td class="num sc">74.0</td> <td class="num sc">87.1</td> <td class="num sc">68.8</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.5-4b%40q6_k" target="_blank" rel="noopener">qwen3.5-4b@q6_k</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:72%;background:#f87171"></i></span><b>71.5</b></td>
<td class="num">28.9</td>
<td class="num sc">54.8</td> <td class="num sc">73.9</td> <td class="num sc">79.1</td> <td class="num sc">78.1</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ggml-org%2Fgpt-oss-20b-GGUF" target="_blank" rel="noopener">ggml-org/gpt-oss-20b-GGUF</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>71.4</b></td>
<td class="num">142.1</td>
<td class="num sc">72.3</td> <td class="num sc">78.8</td> <td class="num sc">57.6</td> <td class="num sc">76.8</td>
<td class="num updated">2026-10-05</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma4%3A26b-a4b-it-qat" target="_blank" rel="noopener">gemma4:26b-a4b-it-qat</a>
<span class="model-hw">AMD Radeon RX 7800 XT</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>71.3</b></td>
<td class="num">72.2</td>
<td class="num sc">64.8</td> <td class="num sc">75.7</td> <td class="num sc">76.2</td> <td class="num sc">68.5</td>
<td class="num updated">2026-10-09</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2Fgemma-4-31B-it-qat-GGUF%3AUD-Q4_K_XL" target="_blank" rel="noopener">unsloth/gemma-4-31B-it-qat-GGUF:UD-Q4_K_XL</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>71.3</b></td>
<td class="num">49.0</td>
<td class="num sc">66.0</td> <td class="num sc">76.5</td> <td class="num sc">72.7</td> <td class="num sc">70.1</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-27B-4bit" target="_blank" rel="noopener">Qwen3.8-27B-4bit</a>
<span class="model-hw">Apple M4 Pro</span></div></td>
<td class="num vram">48 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>71.1</b></td>
<td class="num">28.9</td>
<td class="num sc">53.5</td> <td class="num sc">66.6</td> <td class="num sc">84.7</td> <td class="num sc">79.7</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=huihui-ai%2FHuihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF" target="_blank" rel="noopener">huihui-ai/Huihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.9</b></td>
<td class="num">58.6</td>
<td class="num sc">67.3</td> <td class="num sc">68.3</td> <td class="num sc">86.8</td> <td class="num sc">61.4</td>
<td class="num updated">2026-11-01</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL" target="_blank" rel="noopener">NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.9</b></td>
<td class="num">46.3</td>
<td class="num sc">62.6</td> <td class="num sc">81.1</td> <td class="num sc">77.6</td> <td class="num sc">62.3</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K" target="_blank" rel="noopener">NVIDIA-Nemotron-3.5-Bolt-E4B-Q6_K</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.9</b></td>
<td class="num">42.0</td>
<td class="num sc">61.5</td> <td class="num sc">79.6</td> <td class="num sc">80.7</td> <td class="num sc">61.5</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gpt-oss-120b-Sonnet-Reasoning-Distilled" target="_blank" rel="noopener">gpt-oss-120b-Sonnet-Reasoning-Distilled</a>
<span class="model-hw">Apple M4 Max</span></div></td>
<td class="num vram">128 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.8</b></td>
<td class="num">82.3</td>
<td class="num sc">67.9</td> <td class="num sc">74.7</td> <td class="num sc">70.6</td> <td class="num sc">69.9</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=google%2Fgemma-4-26b-a4b" target="_blank" rel="noopener">google/gemma-4-26b-a4b</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.7</b></td>
<td class="num">114.1</td>
<td class="num sc">67.8</td> <td class="num sc">78.8</td> <td class="num sc">75.7</td> <td class="num sc">60.4</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=huihui-ai%2FHuihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated-MTP-GGUF" target="_blank" rel="noopener">huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated-MTP-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:71%;background:#f87171"></i></span><b>70.7</b></td>
<td class="num">51.7</td>
<td class="num sc">69.5</td> <td class="num sc">68.5</td> <td class="num sc">74.4</td> <td class="num sc">70.3</td>
<td class="num updated">2026-10-04</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2Fgemma-4-26B-A4B-it-GGUF%3AUD-Q5_K_XL" target="_blank" rel="noopener">unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:70%;background:#f87171"></i></span><b>70.2</b></td>
<td class="num">67.1</td>
<td class="num sc">64.2</td> <td class="num sc">78.1</td> <td class="num sc">78.7</td> <td class="num sc">60.0</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=peculiar-ragdoll%2FSharp-Spark-X2.5-4B-GGUF" target="_blank" rel="noopener">peculiar-ragdoll/Sharp-Spark-X2.5-4B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:70%;background:#f87171"></i></span><b>69.6</b></td>
<td class="num">91.2</td>
<td class="num sc">61.9</td> <td class="num sc">79.5</td> <td class="num sc">67.1</td> <td class="num sc">69.8</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma4-26b-opus-distill" target="_blank" rel="noopener">gemma4-26b-opus-distill</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:70%;background:#f87171"></i></span><b>69.6</b></td>
<td class="num">63.0</td>
<td class="num sc">59.6</td> <td class="num sc">71.1</td> <td class="num sc">78.5</td> <td class="num sc">69.1</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma4-26b-uncensored-mtp" target="_blank" rel="noopener">gemma4-26b-uncensored-mtp</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:70%;background:#f87171"></i></span><b>69.6</b></td>
<td class="num">58.3</td>
<td class="num sc">64.7</td> <td class="num sc">67.7</td> <td class="num sc">80.6</td> <td class="num sc">65.7</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma4-26b-a4b-qat" target="_blank" rel="noopener">gemma4-26b-a4b-qat</a>
<span class="model-hw">NVIDIA RTX A2000 12GB + NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:69%;background:#f87171"></i></span><b>69.2</b></td>
<td class="num">63.1</td>
<td class="num sc">65.0</td> <td class="num sc">69.4</td> <td class="num sc">75.7</td> <td class="num sc">66.5</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FNVIDIA-Nemotron-3.5-Bolt-E4B-GGUF" target="_blank" rel="noopener">unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:69%;background:#f87171"></i></span><b>69.0</b></td>
<td class="num">121.1</td>
<td class="num sc">57.8</td> <td class="num sc">85.1</td> <td class="num sc">70.3</td> <td class="num sc">62.7</td>
<td class="num updated">2026-10-28</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL-DFLASH-NVFP4" target="_blank" rel="noopener">NVIDIA-Nemotron-3.5-Bolt-E4B-UD-Q4_K_XL-DFLASH-NVFP4</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:69%;background:#f87171"></i></span><b>69.0</b></td>
<td class="num">61.1</td>
<td class="num sc">63.0</td> <td class="num sc">77.8</td> <td class="num sc">75.3</td> <td class="num sc">60.0</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ornith-ai%2FOrnith-1.5-9B-GGUF%3AQ4_K_M" target="_blank" rel="noopener">ornith-ai/Ornith-1.5-9B-GGUF:Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:69%;background:#f87171"></i></span><b>68.7</b></td>
<td class="num">75.2</td>
<td class="num sc">68.7</td> <td class="num sc">39.7</td> <td class="num sc">86.5</td> <td class="num sc">80.0</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=IFM%2FK2-Horizon-7B-GGUF" target="_blank" rel="noopener">IFM/K2-Horizon-7B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:68%;background:#f87171"></i></span><b>68.1</b></td>
<td class="num">69.5</td>
<td class="num sc">25.8</td> <td class="num sc">86.9</td> <td class="num sc">77.4</td> <td class="num sc">82.4</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=JetBrains%2FMellum2.1-12B-A2.5B-Thinking-GGUF%3AMXFP4_MOE" target="_blank" rel="noopener">JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF:MXFP4_MOE</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">9</td>
<td class="num"><span class="qbar"><i style="width:68%;background:#f87171"></i></span><b>68.0</b></td>
<td class="num">168.0</td>
<td class="num sc">58.4</td> <td class="num sc">72.7</td> <td class="num sc">71.8</td> <td class="num sc">69.1</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=mellum2.1-12b-a2.5b-thinking" target="_blank" rel="noopener">mellum2.1-12b-a2.5b-thinking</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:68%;background:#f87171"></i></span><b>67.9</b></td>
<td class="num">97.1</td>
<td class="num sc">52.6</td> <td class="num sc">75.5</td> <td class="num sc">69.5</td> <td class="num sc">74.0</td>
<td class="num updated">2026-10-09</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=JetBrains%2FMellum2.1-12B-A2.5B-Thinking-GGUF%3AQ4_K_M" target="_blank" rel="noopener">JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF:Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">8</td>
<td class="num"><span class="qbar"><i style="width:67%;background:#f87171"></i></span><b>67.1</b></td>
<td class="num">132.9</td>
<td class="num sc">54.9</td> <td class="num sc">71.5</td> <td class="num sc">70.6</td> <td class="num sc">71.3</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Jan-v3.5-4B-Q4_K_XL" target="_blank" rel="noopener">Jan-v3.5-4B-Q4_K_XL</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:67%;background:#f87171"></i></span><b>67.1</b></td>
<td class="num">73.4</td>
<td class="num sc">40.7</td> <td class="num sc">81.6</td> <td class="num sc">86.0</td> <td class="num sc">60.0</td>
<td class="num updated">2026-10-27</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2Fgemma-4-E4B-it-qat-GGUF" target="_blank" rel="noopener">unsloth/gemma-4-E4B-it-qat-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 3070 Ti</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:67%;background:#f87171"></i></span><b>67.0</b></td>
<td class="num">176.2</td>
<td class="num sc">50.7</td> <td class="num sc">79.3</td> <td class="num sc">74.1</td> <td class="num sc">63.8</td>
<td class="num updated">2026-10-04</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=LFM2.5-24B-A2B-MLX-8bit" target="_blank" rel="noopener">LFM2.5-24B-A2B-MLX-8bit</a>
<span class="model-hw">Apple M5 Max</span></div></td>
<td class="num vram">64 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:67%;background:#f87171"></i></span><b>67.0</b></td>
<td class="num">122.9</td>
<td class="num sc">66.4</td> <td class="num sc">51.5</td> <td class="num sc">77.9</td> <td class="num sc">72.1</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma-4-E4B-it-Q4_0" target="_blank" rel="noopener">gemma-4-E4B-it-Q4_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:67%;background:#f87171"></i></span><b>66.6</b></td>
<td class="num">88.5</td>
<td class="num sc">55.2</td> <td class="num sc">71.1</td> <td class="num sc">74.0</td> <td class="num sc">66.4</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Mellum2-12B-A2.5B-Instruct-MXFP4_MOE" target="_blank" rel="noopener">Mellum2-12B-A2.5B-Instruct-MXFP4_MOE</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:66%;background:#f87171"></i></span><b>66.2</b></td>
<td class="num">86.6</td>
<td class="num sc">59.1</td> <td class="num sc">72.6</td> <td class="num sc">72.9</td> <td class="num sc">60.2</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Qwen3.8-4B-Q5_K_M" target="_blank" rel="noopener">Qwen3.8-4B-Q5_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:66%;background:#f87171"></i></span><b>65.5</b></td>
<td class="num">52.7</td>
<td class="num sc">46.1</td> <td class="num sc">64.7</td> <td class="num sc">77.6</td> <td class="num sc">73.7</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=unsloth%2FGLM-4.7-Flash-GGUF%3AUD-Q4_K_XL" target="_blank" rel="noopener">unsloth/GLM-4.7-Flash-GGUF:UD-Q4_K_XL</a>
<span class="model-hw">AMD Radeon AI PRO R9700</span></div></td>
<td class="num vram">32 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:65%;background:#f87171"></i></span><b>65.3</b></td>
<td class="num">113.7</td>
<td class="num sc">59.7</td> <td class="num sc">69.9</td> <td class="num sc">82.6</td> <td class="num sc">49.0</td>
<td class="num updated">2026-10-17</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=JetBrains%2FMellum2.1-12B-A2.5B-Thinking-GGUF" target="_blank" rel="noopener">JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:65%;background:#f87171"></i></span><b>64.8</b></td>
<td class="num">125.7</td>
<td class="num sc">56.5</td> <td class="num sc">59.8</td> <td class="num sc">72.5</td> <td class="num sc">70.3</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Ministral-3-8B-Instruct-2512-Q4_K_M" target="_blank" rel="noopener">Ministral-3-8B-Instruct-2512-Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:64%;background:#f87171"></i></span><b>64.4</b></td>
<td class="num">34.9</td>
<td class="num sc">53.1</td> <td class="num sc">61.7</td> <td class="num sc">73.2</td> <td class="num sc">69.4</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Spark-X2.6-4B-Q8_0" target="_blank" rel="noopener">Spark-X2.6-4B-Q8_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">24</td>
<td class="num"><span class="qbar"><i style="width:64%;background:#f87171"></i></span><b>63.5</b></td>
<td class="num">39.4</td>
<td class="num sc">48.9</td> <td class="num sc">76.6</td> <td class="num sc">65.6</td> <td class="num sc">63.0</td>
<td class="num updated">2026-11-01</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=google%2Fgemma-4-e2b" target="_blank" rel="noopener">google/gemma-4-e2b</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:62%;background:#f87171"></i></span><b>62.4</b></td>
<td class="num">105.4</td>
<td class="num sc">45.8</td> <td class="num sc">65.2</td> <td class="num sc">73.4</td> <td class="num sc">65.3</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=bartowski%2FOrnith-1.5-9B-GGUF%3AIQ4_XS" target="_blank" rel="noopener">bartowski/Ornith-1.5-9B-GGUF:IQ4_XS</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:61%;background:#f87171"></i></span><b>61.1</b></td>
<td class="num">70.3</td>
<td class="num sc">52.0</td> <td class="num sc">32.7</td> <td class="num sc">85.4</td> <td class="num sc">74.4</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Spark-X2.6-4B-Q4_K_M" target="_blank" rel="noopener">Spark-X2.6-4B-Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 3060</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:61%;background:#f87171"></i></span><b>61.1</b></td>
<td class="num">69.0</td>
<td class="num sc">34.6</td> <td class="num sc">78.4</td> <td class="num sc">65.1</td> <td class="num sc">66.1</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-4b-distill" target="_blank" rel="noopener">qwen3.8-4b-distill</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:59%;background:#f87171"></i></span><b>59.3</b></td>
<td class="num">32.0</td>
<td class="num sc">31.9</td> <td class="num sc">64.8</td> <td class="num sc">70.6</td> <td class="num sc">69.9</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ministral-3-8b-instruct-2512" target="_blank" rel="noopener">ministral-3-8b-instruct-2512</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:59%;background:#f87171"></i></span><b>59.0</b></td>
<td class="num">8.7</td>
<td class="num sc">39.6</td> <td class="num sc">67.8</td> <td class="num sc">77.7</td> <td class="num sc">50.9</td>
<td class="num updated">2026-10-17</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=LiquidAI%2FLFM3-1.7B-GGUF" target="_blank" rel="noopener">LiquidAI/LFM3-1.7B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:58%;background:#f87171"></i></span><b>57.9</b></td>
<td class="num">251.4</td>
<td class="num sc">28.7</td> <td class="num sc">76.0</td> <td class="num sc">48.8</td> <td class="num sc">78.0</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-9B-gguf%3APTQ1_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-9B-gguf:PTQ1_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:58%;background:#f87171"></i></span><b>57.8</b></td>
<td class="num">150.0</td>
<td class="num sc">28.2</td> <td class="num sc">72.0</td> <td class="num sc">63.4</td> <td class="num sc">67.6</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=inclusionAI%2FLing-3.0-tiny-GGUF" target="_blank" rel="noopener">inclusionAI/Ling-3.0-tiny-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:57%;background:#f87171"></i></span><b>56.7</b></td>
<td class="num">210.6</td>
<td class="num sc">45.0</td> <td class="num sc">72.2</td> <td class="num sc">40.5</td> <td class="num sc">69.0</td>
<td class="num updated">2026-10-29</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-9B-gguf%3APQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-9B-gguf:PQ2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">22</td>
<td class="num"><span class="qbar"><i style="width:56%;background:#f87171"></i></span><b>55.7</b></td>
<td class="num">173.8</td>
<td class="num sc">25.7</td> <td class="num sc">65.0</td> <td class="num sc">66.8</td> <td class="num sc">65.2</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=speakleash%2FBielik-11B-v3.0-Instruct-GGUF%3AQ5_K_M" target="_blank" rel="noopener">speakleash/Bielik-11B-v3.0-Instruct-GGUF:Q5_K_M</a>
<span class="model-hw">AMD Radeon RX 9070/9070 XT/9070 GRE</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:55%;background:#f87171"></i></span><b>55.1</b></td>
<td class="num">58.3</td>
<td class="num sc">23.1</td> <td class="num sc">75.4</td> <td class="num sc">71.5</td> <td class="num sc">50.4</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=peculiar-ragdoll%2FSharp-MiniCPM5-2B-GGUF" target="_blank" rel="noopener">peculiar-ragdoll/Sharp-MiniCPM5-2B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:54%;background:#f87171"></i></span><b>53.9</b></td>
<td class="num">146.4</td>
<td class="num sc">17.6</td> <td class="num sc">72.0</td> <td class="num sc">57.9</td> <td class="num sc">68.1</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-4B-gguf%3AQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-4B-gguf:Q2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:53%;background:#f87171"></i></span><b>52.7</b></td>
<td class="num">222.0</td>
<td class="num sc">20.0</td> <td class="num sc">61.3</td> <td class="num sc">72.4</td> <td class="num sc">57.1</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Aldaris%2FQwen3-8B-Q4_K_M-GGUF%3AQ4_K_M" target="_blank" rel="noopener">Aldaris/Qwen3-8B-Q4_K_M-GGUF:Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 2080 Ti</span></div></td>
<td class="num vram">11 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:53%;background:#f87171"></i></span><b>52.6</b></td>
<td class="num">62.5</td>
<td class="num sc">40.0</td> <td class="num sc">57.9</td> <td class="num sc">55.6</td> <td class="num sc">56.8</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=mistralai%2Fministral-3-3b" target="_blank" rel="noopener">mistralai/ministral-3-3b</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:52%;background:#f87171"></i></span><b>52.2</b></td>
<td class="num">24.8</td>
<td class="num sc">34.4</td> <td class="num sc">52.1</td> <td class="num sc">66.7</td> <td class="num sc">55.5</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=nvidia%2Fnemotron-3-nano-4b" target="_blank" rel="noopener">nvidia/nemotron-3-nano-4b</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:52%;background:#f87171"></i></span><b>52.0</b></td>
<td class="num">78.3</td>
<td class="num sc">16.1</td> <td class="num sc">66.5</td> <td class="num sc">61.9</td> <td class="num sc">63.5</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-9B-gguf%3AQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-9B-gguf:Q2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:52%;background:#f87171"></i></span><b>51.8</b></td>
<td class="num">145.8</td>
<td class="num sc">18.4</td> <td class="num sc">59.7</td> <td class="num sc">71.3</td> <td class="num sc">57.7</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-4B-gguf%3APQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-4B-gguf:PQ2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">13</td>
<td class="num"><span class="qbar"><i style="width:51%;background:#f87171"></i></span><b>51.0</b></td>
<td class="num">246.2</td>
<td class="num sc">22.4</td> <td class="num sc">61.4</td> <td class="num sc">63.5</td> <td class="num sc">56.7</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=gemma-3-4b-it" target="_blank" rel="noopener">gemma-3-4b-it</a>
<span class="model-hw">AMD Radeon Vega Series / Radeon Vega Mobile Series</span></div></td>
<td class="num vram">2 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:51%;background:#f87171"></i></span><b>50.8</b></td>
<td class="num">12.7</td>
<td class="num sc">30.6</td> <td class="num sc">43.7</td> <td class="num sc">78.9</td> <td class="num sc">50.2</td>
<td class="num updated">2026-10-21</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ministral-3-3b-instruct-2512" target="_blank" rel="noopener">ministral-3-3b-instruct-2512</a>
<span class="model-hw">AMD Radeon Vega Series / Radeon Vega Mobile Series</span></div></td>
<td class="num vram">2 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:50%;background:#f87171"></i></span><b>50.4</b></td>
<td class="num">13.1</td>
<td class="num sc">32.7</td> <td class="num sc">51.6</td> <td class="num sc">69.4</td> <td class="num sc">48.0</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.5-4B-super-coder.Q4_0" target="_blank" rel="noopener">qwen3.5-4B-super-coder.Q4_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:50%;background:#f87171"></i></span><b>50.3</b></td>
<td class="num">66.0</td>
<td class="num sc">30.0</td> <td class="num sc">46.4</td> <td class="num sc">69.8</td> <td class="num sc">55.1</td>
<td class="num updated">2026-11-01</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=granite-3.3-8b-instruct" target="_blank" rel="noopener">granite-3.3-8b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:50%;background:#f87171"></i></span><b>49.9</b></td>
<td class="num">13.6</td>
<td class="num sc">15.8</td> <td class="num sc">72.0</td> <td class="num sc">64.6</td> <td class="num sc">47.1</td>
<td class="num updated">2026-10-26</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-8B-gguf%3AQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-8B-gguf:Q2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:48%;background:#f87171"></i></span><b>48.5</b></td>
<td class="num">151.1</td>
<td class="num sc">21.2</td> <td class="num sc">64.0</td> <td class="num sc">67.7</td> <td class="num sc">41.1</td>
<td class="num updated">2026-10-10</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=llama3.1%3Alatest" target="_blank" rel="noopener">llama3.1:latest</a>
<span class="model-hw">NVIDIA GeForce RTX 4080 SUPER</span></div></td>
<td class="num vram">16 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:48%;background:#f87171"></i></span><b>47.9</b></td>
<td class="num">99.1</td>
<td class="num sc">31.6</td> <td class="num sc">61.5</td> <td class="num sc">55.4</td> <td class="num sc">43.3</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen%2Fqwen3-1.7b" target="_blank" rel="noopener">qwen/qwen3-1.7b</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:48%;background:#f87171"></i></span><b>47.9</b></td>
<td class="num">57.5</td>
<td class="num sc">21.3</td> <td class="num sc">67.5</td> <td class="num sc">60.5</td> <td class="num sc">42.2</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=openbmb%2FMiniCPM5-2B-GGUF" target="_blank" rel="noopener">openbmb/MiniCPM5-2B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:47%;background:#f87171"></i></span><b>47.4</b></td>
<td class="num">198.0</td>
<td class="num sc">15.8</td> <td class="num sc">53.0</td> <td class="num sc">44.3</td> <td class="num sc">76.2</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=Llama-3.1-Tulu-3-8B-Q4_K_M" target="_blank" rel="noopener">Llama-3.1-Tulu-3-8B-Q4_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:47%;background:#f87171"></i></span><b>47.3</b></td>
<td class="num">42.2</td>
<td class="num sc">29.1</td> <td class="num sc">47.1</td> <td class="num sc">62.7</td> <td class="num sc">50.3</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=smollm3-3b%40q6_k" target="_blank" rel="noopener">smollm3-3b@q6_k</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:47%;background:#f87171"></i></span><b>46.8</b></td>
<td class="num">33.0</td>
<td class="num sc">19.2</td> <td class="num sc">73.2</td> <td class="num sc">56.2</td> <td class="num sc">38.5</td>
<td class="num updated">2026-10-25</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=llama3.1%3Alatest" target="_blank" rel="noopener">llama3.1:latest</a>
<span class="model-hw">AMD Radeon RX 6600/6600 XT/6600M</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:46%;background:#f87171"></i></span><b>46.0</b></td>
<td class="num">52.9</td>
<td class="num sc">28.9</td> <td class="num sc">58.6</td> <td class="num sc">55.5</td> <td class="num sc">40.9</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=llama3.1%3A8b" target="_blank" rel="noopener">llama3.1:8b</a>
<span class="model-hw">NVIDIA RTX 2000 Ada Generation Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:46%;background:#f87171"></i></span><b>45.5</b></td>
<td class="num">29.1</td>
<td class="num sc">32.5</td> <td class="num sc">55.8</td> <td class="num sc">50.8</td> <td class="num sc">43.0</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=mistralai%2Fministral-3-3b" target="_blank" rel="noopener">mistralai/ministral-3-3b</a>
<span class="model-hw">AMD Radeon Vega Series / Radeon Vega Mobile Series</span></div></td>
<td class="num vram">2 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:46%;background:#f87171"></i></span><b>45.5</b></td>
<td class="num">11.0</td>
<td class="num sc">27.6</td> <td class="num sc">46.6</td> <td class="num sc">60.4</td> <td class="num sc">47.5</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=nvidia%2Fnemotron-3-nano-4b" target="_blank" rel="noopener">nvidia/nemotron-3-nano-4b</a>
<span class="model-hw">NVIDIA GeForce GTX 1050</span></div></td>
<td class="num vram">2 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:42%;background:#f87171"></i></span><b>41.9</b></td>
<td class="num">4.9</td>
<td class="num sc">12.7</td> <td class="num sc">43.6</td> <td class="num sc">72.5</td> <td class="num sc">38.6</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=schematron-3b" target="_blank" rel="noopener">schematron-3b</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:40%;background:#f87171"></i></span><b>39.6</b></td>
<td class="num">37.8</td>
<td class="num sc">8.3</td> <td class="num sc">59.7</td> <td class="num sc">56.6</td> <td class="num sc">33.8</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=hermes-3-llama-3.1-8b" target="_blank" rel="noopener">hermes-3-llama-3.1-8b</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:37%;background:#f87171"></i></span><b>37.3</b></td>
<td class="num">21.2</td>
<td class="num sc">19.4</td> <td class="num sc">43.0</td> <td class="num sc">43.2</td> <td class="num sc">43.5</td>
<td class="num updated">2026-10-17</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=smollm2-1.7b-instruct" target="_blank" rel="noopener">smollm2-1.7b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:36%;background:#f87171"></i></span><b>35.8</b></td>
<td class="num">49.4</td>
<td class="num sc">19.8</td> <td class="num sc">40.9</td> <td class="num sc">49.2</td> <td class="num sc">33.3</td>
<td class="num updated">2026-10-25</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ternary-bonsai-2-2b" target="_blank" rel="noopener">ternary-bonsai-2-2b</a>
<span class="model-hw">NVIDIA GeForce GTX 1050</span></div></td>
<td class="num vram">2 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:35%;background:#f87171"></i></span><b>35.3</b></td>
<td class="num">39.1</td>
<td class="num sc">12.2</td> <td class="num sc">39.7</td> <td class="num sc">58.6</td> <td class="num sc">30.5</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen2.5-coder%3A7b" target="_blank" rel="noopener">qwen2.5-coder:7b</a>
<span class="model-hw">NVIDIA RTX 2000 Ada Generation Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:34%;background:#f87171"></i></span><b>34.3</b></td>
<td class="num">52.1</td>
<td class="num sc">41.2</td> <td class="num sc">20.1</td> <td class="num sc">37.8</td> <td class="num sc">38.2</td>
<td class="num updated">2026-10-31</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=llama-3.2-3b-instruct" target="_blank" rel="noopener">llama-3.2-3b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:34%;background:#f87171"></i></span><b>34.3</b></td>
<td class="num">45.7</td>
<td class="num sc">11.6</td> <td class="num sc">29.9</td> <td class="num sc">57.6</td> <td class="num sc">38.2</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=theprint-12b-moe-3a-q4_k_m" target="_blank" rel="noopener">theprint-12b-moe-3a-q4_k_m</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:33%;background:#f87171"></i></span><b>33.4</b></td>
<td class="num">34.1</td>
<td class="num sc">11.7</td> <td class="num sc">30.0</td> <td class="num sc">51.1</td> <td class="num sc">41.0</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=prism-ml%2FTernary-Bonsai-2-2B-gguf%3APQ2_0" target="_blank" rel="noopener">prism-ml/Ternary-Bonsai-2-2B-gguf:PQ2_0</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">11</td>
<td class="num"><span class="qbar"><i style="width:32%;background:#f87171"></i></span><b>32.4</b></td>
<td class="num">372.4</td>
<td class="num sc">7.6</td> <td class="num sc">46.9</td> <td class="num sc">39.3</td> <td class="num sc">35.7</td>
<td class="num updated">2026-10-16</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=allenai.OLMoE-1B-7B-0125-Instruct.Q5_K_M" target="_blank" rel="noopener">allenai.OLMoE-1B-7B-0125-Instruct.Q5_K_M</a>
<span class="model-hw">NVIDIA GeForce RTX 4060 Laptop GPU</span></div></td>
<td class="num vram">8 GB</td>
<td class="num">10</td>
<td class="num"><span class="qbar"><i style="width:31%;background:#f87171"></i></span><b>31.4</b></td>
<td class="num">185.1</td>
<td class="num sc">8.3</td> <td class="num sc">40.5</td> <td class="num sc">53.5</td> <td class="num sc">23.1</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=IFM%2FK2-Horizon-0.9B-GGUF" target="_blank" rel="noopener">IFM/K2-Horizon-0.9B-GGUF</a>
<span class="model-hw">NVIDIA GeForce RTX 4070</span></div></td>
<td class="num vram">12 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:29%;background:#f87171"></i></span><b>29.4</b></td>
<td class="num">284.4</td>
<td class="num sc">4.3</td> <td class="num sc">46.8</td> <td class="num sc">33.8</td> <td class="num sc">32.6</td>
<td class="num updated">2026-10-15</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen3.8-2b-distill" target="_blank" rel="noopener">qwen3.8-2b-distill</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:28%;background:#f87171"></i></span><b>28.2</b></td>
<td class="num">88.4</td>
<td class="num sc">7.6</td> <td class="num sc">17.1</td> <td class="num sc">46.6</td> <td class="num sc">41.6</td>
<td class="num updated">2026-10-13</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=phi-3.1-mini-4k-instruct" target="_blank" rel="noopener">phi-3.1-mini-4k-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">4</td>
<td class="num"><span class="qbar"><i style="width:28%;background:#f87171"></i></span><b>28.2</b></td>
<td class="num">76.2</td>
<td class="num sc">3.2</td> <td class="num sc">38.3</td> <td class="num sc">48.4</td> <td class="num sc">22.9</td>
<td class="num updated">2026-10-14</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=smollm2-1.7b-instruct" target="_blank" rel="noopener">smollm2-1.7b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:25%;background:#f87171"></i></span><b>25.3</b></td>
<td class="num">123.3</td>
<td class="num sc">7.8</td> <td class="num sc">17.1</td> <td class="num sc">47.3</td> <td class="num sc">29.1</td>
<td class="num updated">2026-10-26</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=luciole-1b-instruct-1.1" target="_blank" rel="noopener">luciole-1b-instruct-1.1</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:22%;background:#f87171"></i></span><b>21.6</b></td>
<td class="num">61.7</td>
<td class="num sc">0.0</td> <td class="num sc">31.3</td> <td class="num sc">37.4</td> <td class="num sc">17.6</td>
<td class="num updated">2026-10-30</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=eurollm-1.7b-instruct" target="_blank" rel="noopener">eurollm-1.7b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:10%;background:#f87171"></i></span><b>10.0</b></td>
<td class="num">145.1</td>
<td class="num sc">0.8</td> <td class="num sc">30.2</td> <td class="num sc">8.5</td> <td class="num sc">0.5</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen2.5-0.5b-instruct-q4_k_m" target="_blank" rel="noopener">qwen2.5-0.5b-instruct-q4_k_m</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:9%;background:#f87171"></i></span><b>8.9</b></td>
<td class="num">289.2</td>
<td class="num sc">3.8</td> <td class="num sc">12.5</td> <td class="num sc">8.0</td> <td class="num sc">11.3</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=qwen2.5-0.5b-nnl-distilled-clean-q4_k_m" target="_blank" rel="noopener">qwen2.5-0.5b-nnl-distilled-clean-q4_k_m</a>
<span class="model-hw">AMD Radeon RX 7900 XTX</span></div></td>
<td class="num vram">24 GB</td>
<td class="num">3</td>
<td class="num"><span class="qbar"><i style="width:7%;background:#f87171"></i></span><b>6.7</b></td>
<td class="num">346.3</td>
<td class="num sc">3.4</td> <td class="num sc">10.1</td> <td class="num sc">7.7</td> <td class="num sc">5.5</td>
<td class="num updated">2026-10-20</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=eurollm-1.7b" target="_blank" rel="noopener">eurollm-1.7b</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">5</td>
<td class="num"><span class="qbar"><i style="width:2%;background:#f87171"></i></span><b>2.0</b></td>
<td class="num">166.5</td>
<td class="num sc">1.0</td> <td class="num sc">2.8</td> <td class="num sc">2.3</td> <td class="num sc">1.8</td>
<td class="num updated">2026-10-12</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ministral-3b-instruct" target="_blank" rel="noopener">ministral-3b-instruct</a>
<span class="model-hw">NVIDIA GeForce RTX 2050</span></div></td>
<td class="num vram">4 GB</td>
<td class="num">6</td>
<td class="num"><span class="qbar"><i style="width:2%;background:#f87171"></i></span><b>1.7</b></td>
<td class="num">49.3</td>
<td class="num sc">0.1</td> <td class="num sc">3.7</td> <td class="num sc">1.9</td> <td class="num sc">1.2</td>
<td class="num updated">2026-10-19</td>
</tr>
<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="https://llm-bench.io/benchmarks?model=ministral-1b-base" target="_blank" rel="noopener">ministral-1b-base</a>
<span class="model-hw">NVIDIA GeForce RTX 3060 Laptop GPU</span></div></td>
<td class="num vram">6 GB</td>
<td class="num">7</td>
<td class="num"><span class="qbar"><i style="width:1%;background:#f87171"></i></span><b>0.6</b></td>
<td class="num">183.6</td>
<td class="num sc">0.0</td> <td class="num sc">2.4</td> <td class="num sc">0.0</td> <td class="num sc">0.0</td>
<td class="num updated">2026-10-26</td>
</tr>
</tbody>
</table>
</div>
<footer>
<span>Updated on: <b id="gen">01 Nov 2026</b></span>
<span>·</span>
<span>Data contains runs of the last <b id="win">30</b> days</span>
<span>·</span>
<span>No model with less than <b id="minr">3</b> runs included</span>
<span>·</span>
<span>Methodology: <a href="https://llm-bench.io/benchmark-methodology" target="_blank" rel="noopener">how we measure</a></span>
<span>·</span>
<span>Dataset on Hugging Face: <a href="https://huggingface.co/datasets/llmbenchio/benchmarks-by-vram" target="_blank" rel="noopener">llmbenchio/benchmarks-by-vram</a></span>
</footer>
</div>
<script>
let DATA=null, sortKey="quality", sortDir=-1, vramFilter=null, query="";
async function init(){
// Table rows + meta are pre-rendered at build time (static view for the
// no-JS org card / crawlers). data.json enables interactive
// filtering + sorting on the live Space page.
try{
DATA = await (await fetch("data.json")).json();
}catch(e){ DATA=null; }
if(!DATA) return;
renderChips(); bindHeader(); bindSearch(); render();
}
function renderChips(){
const vals=[...new Set(DATA.rows.map(r=>r.vram))].sort((a,b)=>a-b);
const box=document.getElementById("vramChips");
const all=document.createElement("div");
all.className="chip active"; all.textContent="All VRAM";
all.onclick=()=>{vramFilter=null;[...box.children].forEach(c=>c.classList.remove("active"));all.classList.add("active");render();};
box.appendChild(all);
for(const v of vals){
const c=document.createElement("div");
c.className="chip"; c.textContent=v+" GB";
c.onclick=()=>{vramFilter=v;[...box.children].forEach(x=>x.classList.remove("active"));c.classList.add("active");render();};
box.appendChild(c);
}
}
function bindHeader(){
document.querySelectorAll("th[data-sort]").forEach(th=>{
th.addEventListener("click",()=>{
const k=th.dataset.sort;
if(sortKey===k) sortDir*=-1; else {sortKey=k;sortDir=(k==="model"?1:-1);}
render();
});
});
}
function bindSearch(){
document.getElementById("search").addEventListener("input",e=>{query=e.target.value.toLowerCase();render();});
}
function qColor(v){ if(v==null)return "var(--bad)"; if(v>=85)return "var(--good)"; if(v>=75)return "var(--mid)"; return "var(--bad)"; }
function esc(s){ return String(s).replace(/[&<>"']/g,c=>({"&":"&amp;","<":"&lt;",">":"&gt;",'"':"&quot;","'":"&#39;"}[c])); }
function render(){
let rows=DATA.rows.slice();
if(vramFilter!=null) rows=rows.filter(r=>r.vram===vramFilter);
if(query) rows=rows.filter(r=>r.model.toLowerCase().includes(query));
rows.sort((a,b)=>{
// scenario values are nested under r.scenarios; other keys are top-level
const val=r=>{ const v=(sortKey in r)?r[sortKey]:r.scenarios?.[sortKey]; return v==null?-1:v; };
let x=val(a), y=val(b);
if(typeof x==="string"||typeof y==="string") return sortDir*String(x).localeCompare(String(y));
return sortDir*(x-y);
});
const head=document.querySelector("thead tr");
[...head.children].forEach(th=>{
const t=th.querySelector(".arrow"); if(t)t.remove();
if(th.dataset.sort===sortKey){const a=document.createElement("span");a.className="arrow";a.textContent=sortDir<0?"▼":"▲";th.appendChild(a);}
});
const tb=document.getElementById("rows");
if(!rows.length){ tb.innerHTML='<tr><td colspan="10" class="empty">No models match this filter.</td></tr>'; return; }
tb.innerHTML=rows.map(r=>{
const link=`https://llm-bench.io/benchmarks?model=${encodeURIComponent(r.model)}`;
const sc=k=>r.scenarios[k]!=null
? `<td class="num sc">${r.scenarios[k].toFixed(1)}</td>`
: `<td class="num sc dim">–</td>`;
const q=r.quality;
const qCell=q!=null
? `<td class="num"><span class="qbar"><i style="width:${Math.max(0,Math.min(100,q))}%;background:${qColor(q)}"></i></span><b>${q.toFixed(1)}</b></td>`
: `<td class="num sc dim">–</td>`;
return `<tr style="cursor:pointer" title="Open on llm-bench.io">
<td><div class="model-cell"><a class="model-name" href="${link}" target="_blank" rel="noopener" onclick="event.stopPropagation()">${esc(r.model)}</a><span class="model-hw">${esc(r.hardware)}</span></div></td>
<td class="num vram">${r.vram} GB</td>
<td class="num">${r.runs}</td>
${qCell}
<td class="num">${r.speed!=null?r.speed.toFixed(1):"–"}</td>
${sc("coding_agent")}${sc("openclaw")}${sc("roleplay")}${sc("research")}
<td class="num updated">${r.updated}</td>
</tr>`;
}).join("");
document.getElementById("meta").innerHTML=
`<span><b>${rows.length}</b> models</span>`+
`<span><b>${DATA.inWindow}</b> submissions in window</span>`+
`<span>of <b>${DATA.totalSubmissions}</b> total on llm-bench.io</span>`;
}
init();
</script>
</body>
</html>