Hugging Face's logo

Buckets:
osolmaz
/
ourmodels-data

Files
xet
osolmaz/ourmodels-data / tweets /graph /concepts.json
osolmaz's picture
osolmaz
download
raw
47.1 kB
{
"generated_by": "scripts/extract-tweet-concepts.mjs",
"docs": {
"2079925889735790629": {
"hash": "3c10079b9912d76d",
"concepts": [
"qwen-model",
"small-language-model",
"model-size"
]
},
"2079936019369959750": {
"hash": "e383033dcbe812a3",
"concepts": [
"laguna-model",
"mixture-of-experts-model",
"open-weight-model",
"ai-coding-agent",
"model-parameter",
"context-window",
"serving-runtime",
"quantization"
]
},
"2079938767545413739": {
"hash": "88f1b8577848378a",
"concepts": [
"laguna-model",
"geforce-rtx-3090",
"graphics-processing-unit",
"serving-runtime",
"tensor-parallelism",
"pipeline-parallelism",
"token-throughput"
]
},
"2079945825204293948": {
"hash": "4d6aba79ef6482e3",
"concepts": [
"laguna-model",
"dgx-spark",
"serving-runtime",
"blackwell-architecture",
"unified-memory",
"quantization",
"speculative-decoding",
"context-window"
]
},
"2079959095025508552": {
"hash": "01933ebed7114517",
"concepts": [
"thinky-model",
"model-card",
"serving-runtime",
"inference-speed",
"apache-license"
]
},
"2079959273363390901": {
"hash": "315adf4c5f7c3a16",
"concepts": [
"image-generation-model",
"qwen-model",
"gpt-model",
"nano-banana-model",
"seedream-model"
]
},
"2079962217559253021": {
"hash": "e16c54a916cb78aa",
"concepts": [
"hugging-face",
"bbc",
"media-censorship"
]
},
"2079978448622370888": {
"hash": "fc381997ef96a33f",
"concepts": [
"nanbeige4-2-3b-model",
"gemma-model",
"small-language-model",
"ai-agent",
"benchmark",
"gguf-format",
"context-window",
"transformer-architecture"
]
},
"2079987476798615770": {
"hash": "b3c98c40c5ec5a3a",
"concepts": [
"solar-open-2-model",
"mixture-of-experts-model",
"model-size",
"context-window",
"ai-agent",
"code-generation",
"tool-calling",
"benchmark"
]
},
"2079989350960398445": {
"hash": "36813aff241951ad",
"concepts": [
"nvidia",
"graphics-processing-unit",
"geforce-rtx-3090",
"gpu-peer-to-peer-communication",
"serving-runtime",
"token-throughput",
"low-bit-quantization",
"tensor-parallelism"
]
},
"2079989602597634085": {
"hash": "1b29f75aced23d62",
"concepts": [
"local-ai",
"model-behavior",
"ai-agent",
"benchmark",
"metal-kernel-optimization"
]
},
"2080020279573541085": {
"hash": "da40581062cfcd34",
"concepts": [
"speculative-decoding",
"qwen-model",
"serving-runtime",
"benchmark",
"ai-hardware",
"open-source-software"
]
},
"2080035604624973852": {
"hash": "577239175e6a783a",
"concepts": [
"ai-coding-agent",
"cache-management",
"context-management",
"local-language-model"
]
},
"2080040479337357524": {
"hash": "28e57575b653dac3",
"concepts": [
"glm-model",
"vision-language-model",
"ai-hardware"
]
},
"2080044212519923849": {
"hash": "17dd8625c8d291bb",
"concepts": [
"laguna-model",
"ai-agent",
"qwen-model",
"hermes-model",
"digital-twin",
"data",
"dgx-spark",
"web-user-interface"
]
},
"2080058490341499158": {
"hash": "5bcfe23794ed1adc",
"concepts": [
"gpt-model",
"laguna-model",
"inference-engine",
"inference-speed"
]
},
"2080063136216797678": {
"hash": "f42b3ae29de8c49c",
"concepts": [
"laguna-model",
"low-bit-quantization",
"model-release"
]
},
"2080066129792303446": {
"hash": "1bf4e55c174d07a7",
"concepts": [
"laguna-model",
"low-bit-quantization",
"model-release",
"model-weight"
]
},
"2080072031689670762": {
"hash": "529a50317b4932d6",
"concepts": [
"fable-model",
"kimi-model",
"model-distillation",
"model-training"
]
},
"2080099061617578319": {
"hash": "0453c3e3b6522e80",
"concepts": [
"benchmark",
"laguna-model",
"low-bit-quantization",
"inference-concurrency",
"token-throughput"
]
},
"2080243186497179725": {
"hash": "141d0071dd1daaa8",
"concepts": [
"laguna-model",
"coding-task",
"deepseek-v4-model",
"large-language-model",
"model-specialization"
]
},
"2080247818103796200": {
"hash": "9a32446b7421533b",
"concepts": [
"ai-hardware",
"graphics-processing-unit",
"geforce-rtx-3090",
"used-hardware-market"
]
},
"2080271193093837304": {
"hash": "e344bf715bdcfe37",
"concepts": [
"world-model",
"nvidia-cosmos3-edge-model",
"locateanything-3b-model",
"object-localization",
"local-inference",
"dgx-spark",
"open-weight-model"
]
},
"2080280591988318434": {
"hash": "6fc7f64854bd94ff",
"concepts": [
"ai-hardware",
"token-throughput",
"benchmark"
]
},
"2080282281088835611": {
"hash": "d182230438e04ac8",
"concepts": [
"homelab",
"television",
"sleep"
]
},
"2080287553521868930": {
"hash": "4fba2f73392d7177",
"concepts": [
"quantization",
"tokenizer",
"model-behavior",
"pretraining",
"post-training"
]
},
"2080307789566722276": {
"hash": "143073efa895f6ad",
"concepts": [
"gemma-model",
"mixture-of-experts-model",
"quantization",
"gguf-format",
"context-window",
"kv-cache",
"token-throughput",
"graphics-processing-unit"
]
},
"2080321061954933212": {
"hash": "2a5e3bb76a9ad373",
"concepts": [
"hermes-model",
"large-language-model",
"local-language-model",
"dgx-spark",
"text-to-speech"
]
},
"2080322016410780113": {
"hash": "b8fb91bc24ce3b7e",
"concepts": [
"sparkdash",
"web-user-interface",
"inference-concurrency",
"token-throughput",
"terminal-benchmark",
"inference-engine"
]
},
"2080343905636159880": {
"hash": "affdb162b7d26e59",
"concepts": [
"data",
"benchmark",
"graphics-processing-unit"
]
},
"2080346077442793871": {
"hash": "515031e4e1be2bd8",
"concepts": [
"qwen-model",
"mixture-of-experts-model",
"expert-pruning",
"quantization",
"benchmark",
"inference-speed",
"graphics-processing-unit",
"model-size"
]
},
"2080370423506587765": {
"hash": "0f49f14a2a9a3c3f",
"concepts": [
"open-weight-model",
"model-release",
"model-size"
]
},
"2080406053020127650": {
"hash": "7e57c6174bc21f17",
"concepts": [
"glm-model",
"dgx-spark",
"token-throughput"
]
},
"2080451992670085224": {
"hash": "ed6d5a178c94d82f",
"concepts": [
"laguna-model",
"model-behavior",
"serving-runtime",
"speculative-decoding",
"inference-concurrency",
"kv-cache",
"context-window",
"inference-speed"
]
},
"2080461343724114146": {
"hash": "8e53e072d4b67713",
"concepts": [
"bonsai-model",
"laguna-model",
"quantization",
"model-size",
"model-behavior"
]
},
"2080481180915257656": {
"hash": "eeb16c6627679b2d",
"concepts": [
"dgx-spark",
"deepseek-v4-model",
"ai-hardware",
"inference-concurrency",
"token-throughput",
"context-window"
]
},
"2080490754053455941": {
"hash": "0ae95f1600790dfb",
"concepts": [
"local-ai",
"local-inference",
"inference-engine"
]
},
"2080562131938726132": {
"hash": "b4e60efb127fc9c1",
"concepts": [
"mixture-of-experts-model",
"expert-pruning",
"low-bit-quantization",
"benchmark",
"inference-engine",
"gguf-format",
"ai-hardware",
"model-behavior"
]
},
"2080580719638671454": {
"hash": "26a5d235e20d47cf",
"concepts": [
"open-source-software",
"local-ai",
"ai-ecosystem"
]
},
"2080603956103750119": {
"hash": "d52aebfce09d66e9",
"concepts": [
"benchmark",
"ling-3-0-flash-model",
"laguna-model"
]
},
"2080643318325617033": {
"hash": "0ba3886728a722e9",
"concepts": [
"web-user-interface",
"speech-to-text",
"text-to-speech",
"voice-activity-detection",
"ai-agent",
"message-queue"
]
},
"2080649294655127692": {
"hash": "2d527a8394756a0f",
"concepts": [
"open-weight-model",
"closed-weight-model",
"competition",
"local-ai"
]
},
"2080649400947200237": {
"hash": "48b1262041e4ea34",
"concepts": [
"context-management",
"context-window",
"qwen-model",
"benchmark",
"reinforcement-learning"
]
},
"2080650105288634480": {
"hash": "1919e986b6364947",
"concepts": [
"open-weight-model",
"open-source-software",
"competition"
]
},
"2080650268929126529": {
"hash": "77790f0a843ddd21",
"concepts": [
"open-weight-model",
"open-source-software",
"local-ai",
"infrastructure"
]
},
"2080653579430539559": {
"hash": "c2c521067dd7cbf0",
"concepts": [
"gemma-model",
"quantization",
"vision-language-model",
"speculative-decoding",
"kv-cache",
"inference-speed",
"context-window",
"graphics-processing-unit"
]
},
"2080660182120480872": {
"hash": "c2b478f7f1c8274e",
"concepts": [
"quantization",
"lossless-quantization",
"low-bit-quantization"
]
},
"2080667820673650991": {
"hash": "1a48589895032d48",
"concepts": [
"benchmark",
"blind-evaluation",
"code-generation",
"code-gallery"
]
},
"2080669190470443122": {
"hash": "082d8047429a939a",
"concepts": [
"open-source-software",
"large-language-model",
"transformer-architecture",
"model-distillation",
"open-research"
]
},
"2080669924708556848": {
"hash": "c9f99fd0518f6f2a",
"concepts": [
"lucebox",
"deepseek-v4-model",
"model-release"
]
},
"2080670347980919016": {
"hash": "f13a69f0c974a7b7",
"concepts": [
"open-weight-model",
"open-source-software",
"competition"
]
},
"2080674421266870587": {
"hash": "b42bb4b330967ac2",
"concepts": [
"dgx-spark",
"ai-hardware",
"power-consumption"
]
},
"2080677941625897014": {
"hash": "855b1c50c9750152",
"concepts": [
"local-language-model",
"glm-model",
"open-weight-model",
"fable-model",
"laguna-model",
"deepseek-v4-model",
"benchmark",
"quantization"
]
},
"2080686780676780259": {
"hash": "f2dceb399e4a146f",
"concepts": [
"inference-engine",
"inference-speed",
"model-behavior"
]
},
"2080688525473886369": {
"hash": "c69635c4abfdca54",
"concepts": [
"llada-model",
"ai-agent",
"diffusion-model",
"mixture-of-experts-model",
"context-window",
"open-weight-model",
"benchmark",
"serving-runtime"
]
},
"2080715989788885284": {
"hash": "e39307d2a5fd5d3e",
"concepts": [
"glm-model",
"ai-hardware",
"dgx-spark",
"graphics-processing-unit",
"vision-language-model",
"model-fine-tuning"
]
},
"2080730149348082124": {
"hash": "25b4f80340fd5855",
"concepts": [
"deepseek-v4-model",
"nvidia",
"graphics-processing-unit",
"modelexpress",
"nvidia-dynamo",
"remote-direct-memory-access",
"weight-distribution",
"cache-management"
]
},
"2080739791327617044": {
"hash": "e8314cb6ba646148",
"concepts": [
"nemotron-model",
"dgx-spark",
"vision-language-model",
"quantization",
"context-window",
"token-throughput",
"model-parameter",
"mixture-of-experts-model"
]
},
"2080746063104008408": {
"hash": "58b0b5649fa87e87",
"concepts": [
"ollama",
"glm-model",
"open-weight-model",
"infrastructure",
"model-pricing"
]
},
"2080756202686792071": {
"hash": "027328dc3bafdb4f",
"concepts": [
"laguna-model",
"ai-coding-agent",
"dgx-spark",
"tool-calling",
"context-window",
"benchmark",
"playwright",
"serving-runtime"
]
},
"2080761159938932769": {
"hash": "3d7d0ec57d6d8c1e",
"concepts": [
"ling-3-0-flash-model",
"dgx-spark",
"nvidia"
]
},
"2080782318738776437": {
"hash": "198737894edeabce",
"concepts": [
"glm-model",
"local-ai",
"moral-reasoning",
"philosophy"
]
},
"2080805059890204707": {
"hash": "ea08fe2b21712496",
"concepts": [
"glm-model",
"dgx-spark",
"local-ai",
"model-size",
"power-consumption",
"ai-hardware"
]
},
"2080852503793729639": {
"hash": "158bee0b762aa134",
"concepts": [
"laguna-model",
"ai-coding-agent",
"tool-calling",
"ai-alignment",
"benchmark",
"dgx-spark",
"model-release"
]
},
"2080853152467681662": {
"hash": "9bd31f48e685ffb4",
"concepts": [
"laguna-model",
"strix-halo-platform",
"graphics-processing-unit",
"quantization",
"model-efficiency",
"benchmark",
"poolside-model"
]
},
"2080868138128646473": {
"hash": "906c008dc2152d96",
"concepts": [
"laguna-model",
"model-size",
"ai-hardware",
"quantization",
"kv-cache",
"context-window",
"ai-agent",
"token-throughput"
]
},
"2080869865326342315": {
"hash": "8bd96dcb5bf4593d",
"concepts": [
"benchmark",
"inference-concurrency",
"token-throughput",
"model-size"
]
},
"2080873852087755048": {
"hash": "8030be1df4cc4f6d",
"concepts": [
"glm-model",
"dgx-spark",
"quantization",
"context-window",
"kv-cache",
"model-efficiency"
]
},
"2080875269846388787": {
"hash": "201b34d3867f0823",
"concepts": [
"glm-model",
"dgx-spark",
"local-ai"
]
},
"2080875417703993364": {
"hash": "e7201a92b7f0931b",
"concepts": [
"ling-3-0-flash-model",
"local-ai",
"local-language-model"
]
},
"2080877420890984716": {
"hash": "0e458ac8c5e940bd",
"concepts": [
"deepseek-v4-model",
"competition",
"model-release"
]
},
"2080895772699078745": {
"hash": "5010eb84258048e0",
"concepts": [
"dgx-spark",
"local-language-model",
"ai-hardware"
]
},
"2080897936632791128": {
"hash": "6586b55138798582",
"concepts": [
"quantization",
"model-selection",
"poolside-model"
]
},
"2080900073106337890": {
"hash": "bb23a753f9addb6e",
"concepts": [
"ds4-model",
"local-ai",
"apple-silicon",
"model-fine-tuning",
"unsloth"
]
},
"2080909250243789094": {
"hash": "2a2b7c5ba8cfc6ad",
"concepts": [
"claude-opus-model",
"claude-sonnet-model",
"fable-model",
"large-language-model",
"gpt-model",
"ai-alignment"
]
},
"2080921055435096316": {
"hash": "18acc03d4091d091",
"concepts": [
"openai",
"open-weight-model",
"open-source-software",
"ai-regulation",
"lobbying",
"anthropic"
]
},
"2080925877919949097": {
"hash": "a9992bdefa2368a9",
"concepts": [
"grok-model",
"web-user-interface",
"command-line-interface",
"ai-agent",
"local-ai"
]
},
"2080926692067979561": {
"hash": "f30357df91b0b7e1",
"concepts": [
"gpt-oss-model",
"open-source-software",
"competition"
]
},
"2080928156740211057": {
"hash": "7c932dd755b7bf70",
"concepts": [
"laguna-model",
"speculative-decoding",
"qwen-model",
"quantization",
"serving-runtime",
"gguf-format",
"strix-halo-platform",
"dgx-spark"
]
},
"2080928171718005011": {
"hash": "eb79d8517e53e6f6",
"concepts": [
"qwen-model",
"quantization",
"serving-runtime",
"dgx-spark",
"blackwell-architecture",
"speculative-decoding",
"gguf-format",
"strix-halo-platform"
]
},
"2080933095126237496": {
"hash": "febcf36d05d785e3",
"concepts": [
"open-weight-model",
"local-ai",
"serving-runtime",
"ai-hardware",
"infrastructure",
"open-source-software",
"nvidia"
]
},
"2080940004960866589": {
"hash": "79129a50a39afc46",
"concepts": [
"ds4-model",
"apple-silicon",
"local-ai",
"inference-engine"
]
},
"2080940443819462842": {
"hash": "f4fd77a8d35e0548",
"concepts": [
"dgx-spark",
"unified-memory",
"model-size",
"quantization",
"context-window",
"small-language-model",
"vision-language-model"
]
},
"2080945098142241139": {
"hash": "f7f526c8e754ace4",
"concepts": [
"claude-opus-model",
"fable-model",
"model-pricing",
"model-efficiency",
"open-weight-model",
"open-source-software",
"competition"
]
},
"2080947920036418000": {
"hash": "6bf4cbf51855bb7a",
"concepts": [
"dgx-spark",
"open-weight-model",
"model-selection"
]
},
"2080950137317380386": {
"hash": "5c9c119b58e81768",
"concepts": [
"bonsai-model",
"qwen-model",
"graphics-processing-unit",
"model-size",
"context-window",
"benchmark",
"quantization",
"serving-runtime"
]
},
"2080952619443347601": {
"hash": "2da0f4b97a33e50f",
"concepts": [
"gpt-oss-model",
"open-weight-model",
"closed-weight-model",
"openai"
]
},
"2080955069755711878": {
"hash": "1613ede2fcf7ebbf",
"concepts": [
"nemotron-model",
"local-ai",
"dgx-spark",
"vision-language-model",
"quantization",
"serving-runtime",
"small-language-model",
"model-size"
]
},
"2080956061729636817": {
"hash": "c4ff06a84488bf58",
"concepts": [
"codex",
"ai-coding-agent",
"local-coding-agent"
]
},
"2080959635968741510": {
"hash": "a4e334036c18bf9e",
"concepts": [
"ai-agent",
"large-language-model",
"llm-management"
]
},
"2080964795784724954": {
"hash": "4a1d3168cf2acb8a",
"concepts": [
"local-ai",
"dgx-spark",
"image-generation-model",
"flux-model",
"unified-memory"
]
},
"2080965873397838276": {
"hash": "f5420f49bdbd8df8",
"concepts": [
"mage-flow",
"mlx-vlm",
"vision-language-model"
]
},
"2080972738978783501": {
"hash": "6102ccf11228d4b7",
"concepts": [
"open-weight-model",
"model-size",
"benchmark",
"model-efficiency",
"model-pricing"
]
},
"2081001466261811522": {
"hash": "91402873ab9e8607",
"concepts": [
"local-language-model",
"terminal-benchmark",
"small-language-model",
"state-of-the-art-model",
"ai-assistant",
"research-bot",
"coding-task",
"partial-credit-scoring"
]
},
"2081031099522674696": {
"hash": "74dce10665c51d14",
"concepts": [
"google",
"open-source-software",
"gemma-model"
]
},
"2081044126917611992": {
"hash": "5a85e88e38ae7406",
"concepts": [
"local-language-model",
"large-language-model",
"model-parameter"
]
}
},
"concepts": {
"ai-agent": {
"name": "AI agent",
"aliases": [
"Hermes Agent",
"agents",
"control agents",
"parallel agents",
"agentic AI",
"agent",
"Mastra AgentController",
"agent controller",
"agentic workflows",
"agentic",
"open agent powerhouse"
]
},
"ai-alignment": {
"name": "AI alignment",
"aliases": [
"alignment",
"guardrail"
]
},
"ai-assistant": {
"name": "AI assistant",
"aliases": [
"assistants"
]
},
"ai-coding-agent": {
"name": "AI coding agent",
"aliases": [
"Codex",
"coding agent",
"agent pipeline",
"autonomous agent",
"OpenCode",
"agentic coding"
]
},
"ai-ecosystem": {
"name": "AI ecosystem",
"aliases": [
"State of Local AI - 2026"
]
},
"ai-hardware": {
"name": "AI hardware",
"aliases": [
"hardware",
"DGX Station",
"RTX Pro 6000s",
"3090s",
"2x3090s",
"QSFP cable",
"B300s",
"B300",
"NVIDIA H200",
"H200"
]
},
"ai-regulation": {
"name": "AI regulation",
"aliases": [
"regulation letter",
"open-weight regulation"
]
},
"anthropic": {
"name": "Anthropic",
"aliases": []
},
"apache-license": {
"name": "Apache License",
"aliases": [
"modified Apache",
"Apache"
]
},
"apple-silicon": {
"name": "Apple silicon",
"aliases": [
"M3 Ultra",
"M3 Ultra 512GB"
]
},
"bbc": {
"name": "BBC",
"aliases": []
},
"benchmark": {
"name": "Benchmark",
"aliases": [
"benchmarks",
"honest test",
"tok/s",
"tooleval69",
"eval scores",
"headline benchmark score",
"held-out coding tests",
"held-out work",
"self-play",
"BFCL-V4",
"SWE-bench Verified",
"SWE-bench Pro",
"SWE-bench Multilingual",
"benchmark performance",
"Aeon Bench",
"long-context benchmarks",
"seven long-context benchmarks",
"benchmark scripts",
"HumanEval+",
"MBPP+",
"BFCL",
"GPQA",
"EvalPlus",
"evals",
"localmaxxing",
"benchmark results",
"HumanEval",
"LiveCodeBench"
]
},
"blackwell-architecture": {
"name": "Blackwell architecture",
"aliases": [
"Blackwell",
"Grace-Blackwell",
"blackwell kernels"
]
},
"blind-evaluation": {
"name": "Blind evaluation",
"aliases": [
"blind test"
]
},
"bonsai-model": {
"name": "Bonsai model",
"aliases": [
"bonsai 27b",
"bonsai"
]
},
"cache-management": {
"name": "Cache management",
"aliases": [
"kernel caches",
"prompt cache",
"cached prefix",
"cache efficiency"
]
},
"claude-opus-model": {
"name": "Claude Opus model",
"aliases": [
"claude opus 5",
"Opus 5",
"Opus"
]
},
"claude-sonnet-model": {
"name": "Claude Sonnet model",
"aliases": [
"Sonnet 5",
"Sonnet"
]
},
"closed-weight-model": {
"name": "Closed-weight model",
"aliases": [
"closed weight models",
"closed weight",
"frontier closed",
"closed scoreboard"
]
},
"code-gallery": {
"name": "Code gallery",
"aliases": []
},
"code-generation": {
"name": "Code generation",
"aliases": [
"model outputs",
"coding"
]
},
"codex": {
"name": "Codex",
"aliases": []
},
"coding-task": {
"name": "Coding task",
"aliases": [
"coding tasks"
]
},
"command-line-interface": {
"name": "Command-line interface",
"aliases": [
"CLI",
"Grok CLI"
]
},
"competition": {
"name": "Competition",
"aliases": [
"competitive",
"American AI leadership",
"will win",
"force competition"
]
},
"context-management": {
"name": "Context management",
"aliases": [
"AGENTS.md",
"62K tokens"
]
},
"context-window": {
"name": "Context window",
"aliases": [
"128k context",
"full window",
"massive context",
"316K context",
"256k context",
"128K native context",
"120k context",
"250k context",
"180k ctx",
"long-context",
"512K+ tokens",
"1M token context",
"100K",
"209K",
"180k context",
"1M-token context",
"1M-token",
"262K context window",
"million tokens",
"1M context"
]
},
"data": {
"name": "Data",
"aliases": [
"live factory sensor data",
"sensor data"
]
},
"deepseek-v4-model": {
"name": "DeepSeek v4 model",
"aliases": [
"Deepseek v4",
"Deepseek",
"DeepSeek-V4 Pro",
"DeepSeek V4 Pro",
"DeepSeek v4 release",
"DeepSeek v4 Flash"
]
},
"dgx-spark": {
"name": "DGX Spark",
"aliases": [
"128gb spark",
"Spark",
"dgx-spark",
"DGX Spark units",
"Sparks",
"4× DGX Spark",
"DGX Sparks",
"single DGX Spark",
"NVIDIA DGX Spark"
]
},
"diffusion-model": {
"name": "Diffusion model",
"aliases": [
"agentic diffusion model",
"diffusion generation"
]
},
"digital-twin": {
"name": "Digital twin",
"aliases": [
"3D model"
]
},
"ds4-model": {
"name": "DS4 model",
"aliases": [
"ds4",
"DeepSeek-V4",
"DS4 Flash"
]
},
"expert-pruning": {
"name": "Expert pruning",
"aliases": [
"pruning experts",
"mass-pruning",
"routing-mass",
"REAP",
"deleting half the experts",
"half the experts"
]
},
"fable-model": {
"name": "Fable model",
"aliases": [
"fable 5",
"Fable 5.1",
"Fable"
]
},
"flux-model": {
"name": "FLUX model",
"aliases": [
"flux"
]
},
"geforce-rtx-3090": {
"name": "GeForce RTX 3090",
"aliases": [
"3090",
"used 3090s",
"3090s"
]
},
"gemma-model": {
"name": "Gemma model",
"aliases": [
"Gemma",
"Gemma 4",
"Gemma 4 26B A4B QAT",
"gemma-4-26B",
"Gemma 4 12B"
]
},
"gguf-format": {
"name": "GGUF format",
"aliases": [
"GGUF",
"GGUFs",
"single-file GGUF",
"single-file GGUFs",
"Q4 GGUF"
]
},
"glm-model": {
"name": "GLM model",
"aliases": [
"GLM",
"GLM-5.2",
"GLM-5.2-Vision",
"GLM 5.2",
"GLM-5.2 with vision"
]
},
"google": {
"name": "Google",
"aliases": []
},
"gpt-model": {
"name": "GPT model",
"aliases": [
"GPT 5.2",
"GPT",
"GPT 5.6 Sol",
"GPT Image 2"
]
},
"gpt-oss-model": {
"name": "GPT-OSS model",
"aliases": [
"GPT-OSS",
"GPT oss"
]
},
"gpu-peer-to-peer-communication": {
"name": "GPU peer-to-peer communication",
"aliases": [
"gpu-to-gpu",
"p2p",
"GPU-to-GPU latency"
]
},
"graphics-processing-unit": {
"name": "Graphics processing unit",
"aliases": [
"RTX 3060 Ti",
"8gb card",
"small gpu",
"gfx1151",
"GPU",
"GPU memory",
"RTX 4060",
"RTX 4090",
"8 GB VRAM",
"3090",
"GPUs",
"3090s"
]
},
"grok-model": {
"name": "Grok model",
"aliases": [
"Grok",
"Grok sessions"
]
},
"hermes-model": {
"name": "Hermes model",
"aliases": [
"Hermes",
"NousResearch Hermes"
]
},
"homelab": {
"name": "Homelab",
"aliases": []
},
"hugging-face": {
"name": "Hugging Face",
"aliases": [
"huggingface"
]
},
"image-generation-model": {
"name": "Image generation model",
"aliases": [
"image model",
"Image Models"
]
},
"inference-concurrency": {
"name": "Inference concurrency",
"aliases": [
"concurrency profile",
"six concurrent chats",
"higher concurrency",
"max-num-seqs",
"seqs=32",
"32 concurrent prompts",
"concurrent prompts",
"8-way concurrency"
]
},
"inference-engine": {
"name": "Inference engine",
"aliases": [
"engine",
"vLLM",
"llamacpp",
"inference implementation"
]
},
"inference-speed": {
"name": "Inference speed",
"aliases": [
"tokens/sec",
"prefill throughput",
"decode speed",
"TTFT",
"tok/s",
"cold prefill",
"50 t/s",
"prefill",
"very fast"
]
},
"infrastructure": {
"name": "Infrastructure",
"aliases": [
"infra",
"civilizational infrastructure",
"intelligence infrastructure"
]
},
"kimi-model": {
"name": "Kimi model",
"aliases": [
"Kimi K3"
]
},
"kv-cache": {
"name": "KV cache",
"aliases": [
"KV pool",
"FP8 KV Cache",
"Q8 KV cache",
"Q4 KV",
"KV pressure",
"KV tokens",
"prefix caching",
"KV quantization"
]
},
"laguna-model": {
"name": "Laguna model",
"aliases": [
"Laguna S 2.1",
"Laguna",
"Laguna S2.1",
"Laguna-S 2.1",
"Laguna-S-2.1",
"Laguna-S-2.1-NVFP4",
"Laguna S-2.1"
]
},
"large-language-model": {
"name": "Large language model",
"aliases": [
"LLM",
"LLMs"
]
},
"ling-3-0-flash-model": {
"name": "Ling-3.0-flash model",
"aliases": [
"Ling-3.0-flash"
]
},
"llada-model": {
"name": "LLaDA model",
"aliases": [
"LLaDA2.2-flash",
"LLaDA2.0",
"LLaDA2.1",
"LLaDA2.2"
]
},
"llm-management": {
"name": "LLM management",
"aliases": [
"manages the LLMs"
]
},
"lobbying": {
"name": "Lobbying",
"aliases": []
},
"local-ai": {
"name": "Local AI",
"aliases": [
"local inference",
"on my own hardware",
"local-first",
"run locally",
"locally deployable",
"run intelligence systems",
"run yourself",
"own their stack",
"run models locally"
]
},
"local-coding-agent": {
"name": "Local coding agent",
"aliases": [
"codex at home"
]
},
"local-inference": {
"name": "Local inference",
"aliases": [
"running locally"
]
},
"local-language-model": {
"name": "Local language model",
"aliases": [
"local LLM",
"local models",
"local model",
"LLM",
"local LLMs"
]
},
"locateanything-3b-model": {
"name": "LocateAnything-3B model",
"aliases": [
"LocateAnything-3B"
]
},
"lossless-quantization": {
"name": "Lossless quantization",
"aliases": [
"lossless quant"
]
},
"low-bit-quantization": {
"name": "Low-bit quantization",
"aliases": [
"below 4-bit",
"4-bit",
"INT4",
"requantizing",
"NVFP4",
"AWQ"
]
},
"lucebox": {
"name": "LuceBox",
"aliases": []
},
"mage-flow": {
"name": "Mage Flow",
"aliases": []
},
"media-censorship": {
"name": "Media censorship",
"aliases": [
"can't have that published"
]
},
"message-queue": {
"name": "Message queue",
"aliases": [
"queued message",
"queue/steer messages"
]
},
"metal-kernel-optimization": {
"name": "Metal kernel optimization",
"aliases": [
"Metal Kernel optimizations",
"Metal Kernel"
]
},
"mixture-of-experts-model": {
"name": "Mixture-of-experts model",
"aliases": [
"moe",
"Mixture-of-Experts diffusion architecture",
"experts",
"sparse layers",
"routed experts",
"250B MoE"
]
},
"mlx-vlm": {
"name": "MLX-VLM",
"aliases": []
},
"model-behavior": {
"name": "Model behavior",
"aliases": [
"behavioral info",
"reasoning completely fell apart",
"reasoning into a loop",
"naturalistic prose",
"degrading knowledge",
"endless reasoning loops",
"agent crashes",
"long-context pain",
"logits",
"logits are drifting"
]
},
"model-card": {
"name": "Model card",
"aliases": []
},
"model-distillation": {
"name": "Model distillation",
"aliases": [
"distillation issue",
"distillations"
]
},
"model-efficiency": {
"name": "Model efficiency",
"aliases": [
"efficiency",
"efficiency alone",
"41.4 tok/s peak decode",
"40% faster",
"13% less memory"
]
},
"model-fine-tuning": {
"name": "Model fine-tuning",
"aliases": [
"fine-tune",
"fine-tuning",
"further training"
]
},
"model-parameter": {
"name": "Model parameter",
"aliases": [
"large parameters",
"30b total",
"3b active",
"118B params",
"8B active"
]
},
"model-pricing": {
"name": "Model pricing",
"aliases": [
"pricing",
"half the cost",
"half the price",
"cheaper",
"Max plan",
"Pro plan"
]
},
"model-release": {
"name": "Model release",
"aliases": [
"launch",
"model version",
"revision 0761412",
"release",
"new version"
]
},
"model-selection": {
"name": "Model selection",
"aliases": [
"you pick what i run next",
"run next",
"what quant"
]
},
"model-size": {
"name": "Model size",
"aliases": [
"5T model",
"2.8T",
"half the size",
"30b multimodal model",
"bigger numbers",
"27b model",
"27b",
"33gb footprint",
"27 and 35b models",
"118B-A8B",
"744B-parameter",
"744B",
"8.5B active",
"27B generalist",
"35B",
"size-matched",
"250B",
"15B active",
"small"
]
},
"model-specialization": {
"name": "Model specialization",
"aliases": [
"specialized for coding"
]
},
"model-training": {
"name": "Model training",
"aliases": [
"being trained"
]
},
"model-weight": {
"name": "Model weight",
"aliases": [
"old weights"
]
},
"modelexpress": {
"name": "ModelExpress",
"aliases": [
"NVIDIA ModelExpress"
]
},
"moral-reasoning": {
"name": "Moral reasoning",
"aliases": [
"morality"
]
},
"nanbeige4-2-3b-model": {
"name": "Nanbeige4.2-3B model",
"aliases": [
"Nanbeige4.2-3B",
"Nanbeige"
]
},
"nano-banana-model": {
"name": "Nano Banana model",
"aliases": [
"Nano Banana 2"
]
},
"nemotron-model": {
"name": "Nemotron model",
"aliases": [
"nemotron 3 nano omni"
]
},
"nvidia": {
"name": "NVIDIA",
"aliases": [
"NVIDIAAI",
"nvidia driver"
]
},
"nvidia-cosmos3-edge-model": {
"name": "NVIDIA Cosmos3-Edge model",
"aliases": [
"Cosmos3-Edge"
]
},
"nvidia-dynamo": {
"name": "NVIDIA Dynamo",
"aliases": []
},
"object-localization": {
"name": "Object localization",
"aliases": [
"localized",
"localize"
]
},
"ollama": {
"name": "Ollama",
"aliases": [
"Ollama's cloud"
]
},
"open-research": {
"name": "Open research",
"aliases": []
},
"open-source-software": {
"name": "Open-source software",
"aliases": [
"open source",
"open front",
"built in the open",
"oss",
"Opensource AI",
"open source will win",
"OSS AI",
"Speculators library"
]
},
"open-weight-model": {
"name": "Open-weight model",
"aliases": [
"open-weight models",
"Open-weight",
"open weight",
"open models",
"open weights",
"open alternative",
"frontier-level open models",
"open weights and code",
"open-weight providers",
"open standards",
"frontier open"
]
},
"openai": {
"name": "OpenAI",
"aliases": [
"OAI"
]
},
"partial-credit-scoring": {
"name": "Partial credit scoring",
"aliases": [
"partial scoring"
]
},
"philosophy": {
"name": "Philosophy",
"aliases": []
},
"pipeline-parallelism": {
"name": "Pipeline parallelism",
"aliases": [
"pipeline style"
]
},
"playwright": {
"name": "Playwright",
"aliases": [
"playwright.config.js"
]
},
"poolside-model": {
"name": "Poolside model",
"aliases": [
"poolside one",
"poolside",
"poolsides"
]
},
"post-training": {
"name": "Post-training",
"aliases": []
},
"power-consumption": {
"name": "Power consumption",
"aliases": [
"300 watts",
"electricity bills"
]
},
"pretraining": {
"name": "Pretraining",
"aliases": []
},
"quantization": {
"name": "Quantization",
"aliases": [
"nvfp4",
"1bit",
"1bit crush",
"Q1_0",
"footprint at q8",
"mxfp4",
"ROCmFP4",
"quant",
"4-bit",
"FP8",
"ROCmFP4-fast quant",
"Q quants",
"Q4km",
"q8 quantization",
"1-bit",
"2-bit quantization",
"low-bit quantization",
"lossless quantization",
"QAT",
"Q8 KV",
"Q4 KV",
"ternary quantized",
"quants",
"quantize",
"nvfp4 quant",
"quantized models"
]
},
"qwen-model": {
"name": "Qwen model",
"aliases": [
"qwen 3.6 27b",
"Qwen3.6",
"Qwen3.6-27B",
"Qwen3.6-35B-A3B",
"Qwen3-4B",
"Qwen3-8B",
"Qwen3-30B-A3B",
"Qwen3.6 35B A3B",
"Qwen3.8 Max Preview",
"Qwen",
"Qwen3.5-397B-A17B",
"Qwen3.5",
"Qwen Image 3.0",
"Qwen 3.6"
]
},
"reinforcement-learning": {
"name": "Reinforcement learning",
"aliases": [
"RL training sets",
"64K training"
]
},
"remote-direct-memory-access": {
"name": "Remote direct memory access",
"aliases": [
"GPU-to-GPU RDMA",
"RDMA"
]
},
"research-bot": {
"name": "Research bot",
"aliases": [
"research bots"
]
},
"seedream-model": {
"name": "Seedream model",
"aliases": [
"Seedream 5.0 Pro"
]
},
"serving-runtime": {
"name": "Serving runtime",
"aliases": [
"vllm",
"llama.cpp",
"llama-server",
"llama-cli",
"stack to serve them",
"rocmfp4-llama",
"llama-cpp",
"native harness",
"custom diffusion/MoE runtime",
"FlashInfer",
"actual vllm load",
"SGLang"
]
},
"sleep": {
"name": "Sleep",
"aliases": [
"falls asleep"
]
},
"small-language-model": {
"name": "Small language model",
"aliases": [
"smaller models",
"small models",
"small model",
"3B model"
]
},
"solar-open-2-model": {
"name": "Solar Open 2 model",
"aliases": [
"Solar Open 2"
]
},
"sparkdash": {
"name": "SparkDash",
"aliases": []
},
"speculative-decoding": {
"name": "Speculative decoding",
"aliases": [
"DFlash",
"baked DFlash",
"drafts",
"MTP",
"self speculative decoding",
"draft model",
"spec-decode",
"draft sampling",
"speculator",
"draft tokens",
"dflash drafter"
]
},
"speech-to-text": {
"name": "Speech-to-text",
"aliases": [
"STT"
]
},
"state-of-the-art-model": {
"name": "State-of-the-art model",
"aliases": [
"sota models"
]
},
"strix-halo-platform": {
"name": "Strix Halo platform",
"aliases": [
"gfx1151",
"Strix-class AMD",
"Strix Halo"
]
},
"television": {
"name": "Television",
"aliases": []
},
"tensor-parallelism": {
"name": "Tensor parallelism",
"aliases": [
"tp4",
"tp2",
"tensor parallel"
]
},
"terminal-benchmark": {
"name": "Terminal benchmark",
"aliases": [
"terminal bench",
"TB2.1",
"DecodeBench",
"bench results"
]
},
"text-to-speech": {
"name": "Text-to-speech",
"aliases": [
"TTS"
]
},
"thinky-model": {
"name": "Thinky model",
"aliases": [
"Thinky"
]
},
"token-throughput": {
"name": "Token throughput",
"aliases": [
"4000tps",
"1k tokens/second",
"1300 tok/s prefill",
"100+ tok/s aggregate",
"22 tok/s",
"tok/s",
"server tok/s",
"tokens/sec",
"t/s",
"tps",
"aggregate throughput",
"aggregating throughput",
"throughput"
]
},
"tokenizer": {
"name": "Tokenizer",
"aliases": []
},
"tool-calling": {
"name": "Tool calling",
"aliases": [
"tool calls",
"tool-call",
"tool use"
]
},
"transformer-architecture": {
"name": "Transformer architecture",
"aliases": [
"transformer revolution",
"Looped Transformer",
"transformer layers"
]
},
"unified-memory": {
"name": "Unified memory",
"aliases": [
"128gb of unified memory",
"128gb"
]
},
"unsloth": {
"name": "Unsloth",
"aliases": [
"Unsloth guides",
"Unsloth Studio"
]
},
"used-hardware-market": {
"name": "Used hardware market",
"aliases": [
"used 3090s"
]
},
"vision-language-model": {
"name": "Vision-language model",
"aliases": [
"VLM",
"multimodal model",
"images or documents",
"open multimodal model",
"GLM-5.2-Vision",
"vision",
"multimodal",
"vision projector",
"mmproj",
"GLM-5.2 with vision"
]
},
"voice-activity-detection": {
"name": "Voice activity detection",
"aliases": [
"VAD"
]
},
"web-user-interface": {
"name": "Web user interface",
"aliases": [
"Web UI",
"dashboard",
"voice ui",
"interactive voice ui",
"browser control room"
]
},
"weight-distribution": {
"name": "Weight distribution",
"aliases": [
"weight movement"
]
},
"world-model": {
"name": "World model",
"aliases": [
"world models"
]
}
}
}

Xet Storage Details

Size:
47.1 kB
·
Xet hash:
ef4b3362198f627957217d991528741a6dbbf4941476d2df34cfb17e3c542f34

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.