NakliTechie commited on
Commit
b17e86d
·
1 Parent(s): f5a958f

Sync from GitHub 2026-10-24T09:23:34Z

Browse files
Files changed (1) hide show
  1. index.html +50 -12
index.html CHANGED
@@ -2689,6 +2689,7 @@
2689
  <li><strong>LFM2.5 24B-A2B &middot; SSD-streamed (experimental)</strong> (~13.6 GB) &mdash; text, reasoning-only. Liquid AI's LFM2.5-24B-A2B (released 2026-10-13): a 24B mixture-of-experts (64 experts, 4 active per token, about 2.4B parameters per token) with short-convolution and attention layers, from Liquid's own 4-bit GGUF. Short-convolution and attention layers, routers and KV cache stay on the GPU while the 2,432 routed experts stream from on-device storage (OPFS) as the router picks them. The first load copies the model into this browser's storage once (~13.6 GB of disk). About 16 tokens/s on an M4 Pro with 24 GB, and prompts are read in chunks at about 50 tokens/s. Greedy output matches llama.cpp (8 of 8 prompts). It always reasons before it answers, so every reply opens with its reasoning. It can use tools (calculator, time, memory, reminders): it called the right tool on 6 of 6 test prompts and answered 5 of 5 plain questions without one.</li>
2690
  <li><strong>Flint 40B-A2.5B &middot; SSD-streamed (experimental)</strong> (~26 GB) &mdash; text, coding, reasoning. Recursive's Flint (released 2026-10-21, Apache-2.0): a 39.8B mixture-of-experts coding model (256 experts, 6 active per token, about 2.5B parameters per token), from Recursive's own 4-bit (Q4_1) GGUF, whose card asks for 48 GB of memory under llama.cpp. Attention, routers, the token table and KV cache stay on the GPU while the 12,032 routed experts stream from on-device storage (OPFS) as the router picks them. The first load copies the model into this browser's storage once (~26 GB of disk). About 8.7 tokens/s on an M4 Pro with 24 GB, and prompts are read at about 27 tokens/s. Its output stays as close to llama.cpp's Metal backend as llama.cpp's own CPU backend does. It reasons before it answers, so replies open with its reasoning. Tools are off for now.</li>
2691
  <li><strong>Ling 3.0 Flash 124B-A5B &middot; via anyformer (experimental)</strong> (~81 GB) &mdash; text, reasoning. inclusionAI's Ling 3.0 Flash (released 2026-08-02, MIT): a 124B mixture-of-experts (512 experts, 8 active per token, about 5.1B parameters per token) whose layers mix linear attention (KDA) and latent attention (MLA), from inclusionAI's own 4-bit Q4_K_M GGUF (77 GB in two parts). anyformer loads it on diskformer's engine for it, which keeps the attention, routers and caches on the GPU (about 2.5 GB plus a 4 GB expert cache; it needs about 7 GB of GPU memory) and streams the 20,480 routed experts from on-device storage (OPFS). The first load copies the model into this browser's storage once (~81 GB of disk). About 5.5 tokens/s on an M4 Pro with 24 GB, and prompts are read at about 5 tokens/s. Flip <strong>Show reasoning</strong> for its thinking. Tools are off for now.</li>
 
2692
  <li><strong>GLM-4.7-Flash 30B-A3B &middot; via anyformer (experimental)</strong> (~18 GB) &mdash; text, reasoning. Z.ai's GLM-4.7-Flash, a 30B mixture-of-experts with about 3B parameters per token, from unsloth's 4-bit GGUF (18.3 GB). LocalMind has no engine of its own for it: anyformer picks diskformer's llama.cpp path, which runs llama.cpp's WebGPU build in a worker and pages the experts in from on-device storage (OPFS). The first load copies the file into this browser's storage once. Flip <strong>Show reasoning</strong> for its thinking. Tools are off for now.</li>
2693
  <li><strong>Qwen3 4B &middot; FlashMap speculative (experimental)</strong> (~5.2 GB) &mdash; text, reasoning. Qwen's Qwen3-4B (Q8_0 GGUF, 4.3 GB) with FlashMap, a speculative-decoding drafter by Shen et al. (arXiv:2610.24365; 0.95 GB) that proposes 16 tokens in one step for the model to check in one pass. You get exactly the tokens the model would write on its own, sooner: on an M4 Pro about 2.1&ndash;2.2&times; on maths and 1.2&ndash;1.5&times; on code. Chat text drafts poorly, so there it falls back to plain decoding (about 0.7&ndash;1.1&times;). Both files are copied into this browser's storage once. Tools are off for now.</li>
2694
  </ul>
@@ -3008,6 +3009,7 @@
3008
  <option value="lfm25-24b-a2b-ssd">LFM2.5 24B-A2B &middot; SSD-streamed, experimental (~13.6 GB)</option>
3009
  <option value="flint-40b-a2b-ssd">Flint 40B-A2.5B &middot; SSD-streamed, experimental, coding (~26 GB)</option>
3010
  <option value="ling-3-flash-anyformer">Ling 3.0 Flash 124B-A5B &middot; via anyformer, experimental (~81 GB)</option>
 
3011
  <option value="glm-4.7-flash-anyformer">GLM-4.7-Flash 30B-A3B &middot; via anyformer, experimental (~18 GB)</option>
3012
  <option value="qwen3-4b-flashmap">Qwen3 4B &middot; FlashMap speculative decoding, experimental (~5.2 GB)</option>
3013
  </select>
@@ -8017,7 +8019,7 @@ const engineFetch = async (url, init = {}) => {
8017
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy
8018
  // of the engine (Gemma4Mobile) and, with ple, builds the OPFS per-layer-embedding table itself. The worker then drives
8019
  // session.engine exactly as it drove its own engine. Pinned to the same anyformer as anyformer_models.js.
8020
- const ANYFORMER_URL = 'https://cdn.jsdelivr.net/npm/anyformer@0.4.4/index.js';
8021
 
8022
  let anyformerLoad = null;
8023
  let session = null;
@@ -8162,7 +8164,7 @@ self.onmessage = async (e) => {
8162
  // rolled-in module URLs (moduleUrl()).
8163
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy of
8164
  // the engine (TernaryBonsai2); the worker drives session.engine as before and attaches DFlash / MTP itself.
8165
- const ANYFORMER_URL = 'https://cdn.jsdelivr.net/npm/anyformer@0.4.4/index.js';
8166
  const DFLASH_URL = '__TERNARY_BONSAI2_DFLASH_URL__';
8167
  const MTP_URL = '__TERNARY_BONSAI2_MTP_URL__';
8168
  // MTP speculative decoding (Settings → Models, default OFF; posted as `mtp` + `mtpDepth` on the
@@ -9428,6 +9430,20 @@ self.addEventListener('message', async (e) => {
9428
  contextSize: 4096,
9429
  genConfig: { max_new_tokens: 2048 },
9430
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9431
  'glm-4.7-flash-anyformer': {
9432
  id: 'unsloth/GLM-4.7-Flash-GGUF',
9433
  label: 'GLM-4.7-Flash 30B-A3B · via anyformer (experimental)',
@@ -17795,6 +17811,7 @@ self.addEventListener('message', async (e) => {
17795
  'anyformer-qwen36': { engine: 'anyformer_models.js', exportName: 'Qwen36', label: 'Qwen3.6 35B-A3B (Q8_0, via anyformer)' },
17796
  'anyformer-ling3flash': { engine: 'anyformer_models.js', exportName: 'Ling3Flash', label: 'Ling 3.0 Flash (Q4_K_M, via anyformer)' },
17797
  'anyformer-glm47flash': { engine: 'anyformer_models.js', exportName: 'Glm47Flash', label: 'GLM-4.7-Flash (Q4_K_M, via anyformer)' },
 
17798
  'qwen3dense': { engine: 'qwen35_dense.js', exportName: 'Qwen3Dense', label: 'Qwen3 4B (Q8_0) + FlashMap' },
17799
  'hyv3-ssd': { engine: 'hy_v3_moe_ssd.js', exportName: 'HyV3MoeSsd', label: 'Hy-Memory 30B-A3B (Q8_0)' }, // the memory writer only (WRITER_MODELS)
17800
  'qwen35dense': { engine: 'qwen35_dense.js', exportName: 'Qwen35Dense', label: 'Underdog Saluki 27B (IQ2-mix)' },
@@ -19406,7 +19423,7 @@ self.addEventListener('message', async (e) => {
19406
  const ENGINE_STORES = {
19407
  'transformers': ['transformers-cache'], 'wllama': ['wllama'], 'lfm2-webgpu': ['gguf-v1'],
19408
  'bonsai2-webgpu': ['gguf-cache-v1'], 'gemma4-webgpu': ['safetensors-cache-v1', 'bonsai-pipeline-v1'],
19409
- 'qwen35moe-ssd': ['on disk (OPFS)'], 'minicpmv-ssd': ['on disk (OPFS)'], 'gemma4moe-ssd': ['on disk (OPFS)'], 'lfm2moe-ssd': ['on disk (OPFS)'], 'flint-ssd': ['on disk (OPFS)'], 'anyformer-glm47flash': ['on disk (OPFS)'], 'anyformer-ling3flash': ['on disk (OPFS)'], 'anyformer-qwen36': ['on disk (OPFS)'], 'anyformer-flint': ['on disk (OPFS)'], 'anyformer-lfm25-24b': ['on disk (OPFS)'], 'anyformer-gemma26b': ['on disk (OPFS)'], 'qwen3dense': ['on disk (OPFS)'], 'qwen35dense': ['on disk (OPFS)'],
19410
  };
19411
  function parseSizeBytes(size) {
19412
  const m = /([\d.]+)\s*(GB|MB)/i.exec(String(size || ''));
@@ -24042,32 +24059,45 @@ self.addEventListener('message', async (e) => {
24042
  */
24043
  const THINK_CLOSE = -2;
24044
 
24045
- export const ANYFORMER_URL = 'https://cdn.jsdelivr.net/npm/anyformer@0.4.4/index.js';
24046
 
24047
  // anyformer's progress events → the engine events the SSD worker posts. The llama.cpp path reports 'download' (bytes
24048
  // into OPFS) and 'load' / 'plan'; a hand engine (diskformer back end) reports its own engine events, 'weights' with
24049
- // bytes while it ingests and with kind 'tensors' while it uploads, and 'init'. Every byte event must reach the worker:
24050
- // the page's load watchdog ends a load that reports no progress for 3 minutes (the Ling production check, 2026-10-23).
24051
- function toEngineProgress(e) {
 
 
24052
  if (!e) return null;
24053
  if (e.status === 'download') return { status: 'weights', loaded: e.loaded, total: e.total };
 
 
 
 
 
 
24054
  if (e.status === 'weights') return { status: 'weights', kind: e.kind, loaded: e.loaded, total: e.total };
24055
  if (e.status === 'load' || e.status === 'plan' || e.status === 'init') return { status: 'init' };
24056
  return null;
24057
  }
24058
 
24059
  export class AnyformerModel {
 
 
 
24060
  static repo = null;
24061
  static file = undefined; // a GGUF in the repo; undefined lets anyformer pick (the file diskformer checked)
24062
  static budget = 4e9; // GPU bytes: the llama.cpp path plans its expert slots from it; for a hand engine, the fit check
24063
  static poolBytes = undefined; // a hand engine's expert cache (LocalMind's own engines ran with 4 GiB)
24064
 
24065
  static async load(_id, opts = {}) {
24066
- const { load } = await import(ANYFORMER_URL);
24067
- const session = await load(this.repo, {
24068
- budget: opts.budget || this.budget, maxCtx: opts.maxCtx, signal: opts.signal, ...(this.file ? { file: this.file } : {}),
 
 
24069
  ...(this.poolBytes ? { poolBytes: this.poolBytes } : {}), ...(opts.localFile ? { localFile: opts.localFile } : {}),
24070
- onProgress: (e) => { const ev = toEngineProgress(e); if (ev && opts.onProgress) opts.onProgress(ev); },
24071
  });
24072
  return new this(session);
24073
  }
@@ -24150,6 +24180,14 @@ export class Flint40B extends AnyformerModel {
24150
  static poolBytes = 4 * 2 ** 30;
24151
  }
24152
 
 
 
 
 
 
 
 
 
24153
  // LFM2.5 24B-A2B: Liquid AI's QAD-Q4_0 GGUF (13.6 GB) on diskformer's Lfm2MoeSsd hand engine (anyformer v0.4.2+). It
24154
  // always reasons; session.think carries the markers. A 4 GiB expert cache as before.
24155
  export class Lfm25_24B extends AnyformerModel {
@@ -32388,7 +32426,7 @@ const GGUF_FILE = 'LFM2.5-230M-Q4_0.gguf';
32388
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy of
32389
  // the LFM2.5 engine (Lfm2Mobile) and returns a session with the same generate / reset / dispose. Pinned to an exact
32390
  // version; LocalMind's anyformer_models.js pins the same one.
32391
- const ANYFORMER_URL = 'https://cdn.jsdelivr.net/npm/anyformer@0.4.4/index.js';
32392
 
32393
  let anyformerLoad = null;
32394
  let model = null;
 
2689
  <li><strong>LFM2.5 24B-A2B &middot; SSD-streamed (experimental)</strong> (~13.6 GB) &mdash; text, reasoning-only. Liquid AI's LFM2.5-24B-A2B (released 2026-10-13): a 24B mixture-of-experts (64 experts, 4 active per token, about 2.4B parameters per token) with short-convolution and attention layers, from Liquid's own 4-bit GGUF. Short-convolution and attention layers, routers and KV cache stay on the GPU while the 2,432 routed experts stream from on-device storage (OPFS) as the router picks them. The first load copies the model into this browser's storage once (~13.6 GB of disk). About 16 tokens/s on an M4 Pro with 24 GB, and prompts are read in chunks at about 50 tokens/s. Greedy output matches llama.cpp (8 of 8 prompts). It always reasons before it answers, so every reply opens with its reasoning. It can use tools (calculator, time, memory, reminders): it called the right tool on 6 of 6 test prompts and answered 5 of 5 plain questions without one.</li>
2690
  <li><strong>Flint 40B-A2.5B &middot; SSD-streamed (experimental)</strong> (~26 GB) &mdash; text, coding, reasoning. Recursive's Flint (released 2026-10-21, Apache-2.0): a 39.8B mixture-of-experts coding model (256 experts, 6 active per token, about 2.5B parameters per token), from Recursive's own 4-bit (Q4_1) GGUF, whose card asks for 48 GB of memory under llama.cpp. Attention, routers, the token table and KV cache stay on the GPU while the 12,032 routed experts stream from on-device storage (OPFS) as the router picks them. The first load copies the model into this browser's storage once (~26 GB of disk). About 8.7 tokens/s on an M4 Pro with 24 GB, and prompts are read at about 27 tokens/s. Its output stays as close to llama.cpp's Metal backend as llama.cpp's own CPU backend does. It reasons before it answers, so replies open with its reasoning. Tools are off for now.</li>
2691
  <li><strong>Ling 3.0 Flash 124B-A5B &middot; via anyformer (experimental)</strong> (~81 GB) &mdash; text, reasoning. inclusionAI's Ling 3.0 Flash (released 2026-08-02, MIT): a 124B mixture-of-experts (512 experts, 8 active per token, about 5.1B parameters per token) whose layers mix linear attention (KDA) and latent attention (MLA), from inclusionAI's own 4-bit Q4_K_M GGUF (77 GB in two parts). anyformer loads it on diskformer's engine for it, which keeps the attention, routers and caches on the GPU (about 2.5 GB plus a 4 GB expert cache; it needs about 7 GB of GPU memory) and streams the 20,480 routed experts from on-device storage (OPFS). The first load copies the model into this browser's storage once (~81 GB of disk). About 5.5 tokens/s on an M4 Pro with 24 GB, and prompts are read at about 5 tokens/s. Flip <strong>Show reasoning</strong> for its thinking. Tools are off for now.</li>
2692
+ <li><strong>Nemotron 3.5 Bolt E4B &middot; via anyformer (experimental)</strong> (~4.9 GB) &mdash; text. NVIDIA's Nemotron 3.5 Bolt E4B, a dense model with about 4B effective parameters, post-trained from Google's Gemma 4 E4B. It runs on Transformers.js through anyformer, from a community 4-bit ONNX conversion (text only), at about 35 tokens a second on an M4 Pro. Reasoning and tools are off for now.</li>
2693
  <li><strong>GLM-4.7-Flash 30B-A3B &middot; via anyformer (experimental)</strong> (~18 GB) &mdash; text, reasoning. Z.ai's GLM-4.7-Flash, a 30B mixture-of-experts with about 3B parameters per token, from unsloth's 4-bit GGUF (18.3 GB). LocalMind has no engine of its own for it: anyformer picks diskformer's llama.cpp path, which runs llama.cpp's WebGPU build in a worker and pages the experts in from on-device storage (OPFS). The first load copies the file into this browser's storage once. Flip <strong>Show reasoning</strong> for its thinking. Tools are off for now.</li>
2694
  <li><strong>Qwen3 4B &middot; FlashMap speculative (experimental)</strong> (~5.2 GB) &mdash; text, reasoning. Qwen's Qwen3-4B (Q8_0 GGUF, 4.3 GB) with FlashMap, a speculative-decoding drafter by Shen et al. (arXiv:2610.24365; 0.95 GB) that proposes 16 tokens in one step for the model to check in one pass. You get exactly the tokens the model would write on its own, sooner: on an M4 Pro about 2.1&ndash;2.2&times; on maths and 1.2&ndash;1.5&times; on code. Chat text drafts poorly, so there it falls back to plain decoding (about 0.7&ndash;1.1&times;). Both files are copied into this browser's storage once. Tools are off for now.</li>
2695
  </ul>
 
3009
  <option value="lfm25-24b-a2b-ssd">LFM2.5 24B-A2B &middot; SSD-streamed, experimental (~13.6 GB)</option>
3010
  <option value="flint-40b-a2b-ssd">Flint 40B-A2.5B &middot; SSD-streamed, experimental, coding (~26 GB)</option>
3011
  <option value="ling-3-flash-anyformer">Ling 3.0 Flash 124B-A5B &middot; via anyformer, experimental (~81 GB)</option>
3012
+ <option value="nemotron-bolt-e4b-anyformer">Nemotron 3.5 Bolt E4B &middot; via anyformer, experimental (~4.9 GB)</option>
3013
  <option value="glm-4.7-flash-anyformer">GLM-4.7-Flash 30B-A3B &middot; via anyformer, experimental (~18 GB)</option>
3014
  <option value="qwen3-4b-flashmap">Qwen3 4B &middot; FlashMap speculative decoding, experimental (~5.2 GB)</option>
3015
  </select>
 
8019
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy
8020
  // of the engine (Gemma4Mobile) and, with ple, builds the OPFS per-layer-embedding table itself. The worker then drives
8021
  // session.engine exactly as it drove its own engine. Pinned to the same anyformer as anyformer_models.js.
8022
+ const ANYFORMER_URL = 'https://cdn.jsdelivr.net/gh/NakliTechie/anyformer.js@v0.4.6/index.js';
8023
 
8024
  let anyformerLoad = null;
8025
  let session = null;
 
8164
  // rolled-in module URLs (moduleUrl()).
8165
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy of
8166
  // the engine (TernaryBonsai2); the worker drives session.engine as before and attaches DFlash / MTP itself.
8167
+ const ANYFORMER_URL = 'https://cdn.jsdelivr.net/gh/NakliTechie/anyformer.js@v0.4.6/index.js';
8168
  const DFLASH_URL = '__TERNARY_BONSAI2_DFLASH_URL__';
8169
  const MTP_URL = '__TERNARY_BONSAI2_MTP_URL__';
8170
  // MTP speculative decoding (Settings → Models, default OFF; posted as `mtp` + `mtpDepth` on the
 
9430
  contextSize: 4096,
9431
  genConfig: { max_new_tokens: 2048 },
9432
  },
9433
+ // backend:'anyformer-bolt-e4b' → NVIDIA's Nemotron 3.5 Bolt E4B through anyformer, which sends it to transformers.js
9434
+ // (a community q4f16 ONNX, text only). Reasoning off: anyformer's text-only Gemma 4 path renders the template without it.
9435
+ 'nemotron-bolt-e4b-anyformer': {
9436
+ id: 'danyalmalik/NVIDIA-Nemotron-3.5-Bolt-E4B-ONNX-WebGPU',
9437
+ label: 'Nemotron 3.5 Bolt E4B · via anyformer (experimental)',
9438
+ backend: 'anyformer-bolt-e4b',
9439
+ size: '~4.9 GB',
9440
+ type: 'causal',
9441
+ multimodal: false,
9442
+ agentCapable: false,
9443
+ thinking: false,
9444
+ contextSize: 4096,
9445
+ genConfig: { max_new_tokens: 2048 },
9446
+ },
9447
  'glm-4.7-flash-anyformer': {
9448
  id: 'unsloth/GLM-4.7-Flash-GGUF',
9449
  label: 'GLM-4.7-Flash 30B-A3B · via anyformer (experimental)',
 
17811
  'anyformer-qwen36': { engine: 'anyformer_models.js', exportName: 'Qwen36', label: 'Qwen3.6 35B-A3B (Q8_0, via anyformer)' },
17812
  'anyformer-ling3flash': { engine: 'anyformer_models.js', exportName: 'Ling3Flash', label: 'Ling 3.0 Flash (Q4_K_M, via anyformer)' },
17813
  'anyformer-glm47flash': { engine: 'anyformer_models.js', exportName: 'Glm47Flash', label: 'GLM-4.7-Flash (Q4_K_M, via anyformer)' },
17814
+ 'anyformer-bolt-e4b': { engine: 'anyformer_models.js', exportName: 'NemotronBoltE4B', label: 'Nemotron 3.5 Bolt E4B (q4f16 ONNX, via anyformer)' },
17815
  'qwen3dense': { engine: 'qwen35_dense.js', exportName: 'Qwen3Dense', label: 'Qwen3 4B (Q8_0) + FlashMap' },
17816
  'hyv3-ssd': { engine: 'hy_v3_moe_ssd.js', exportName: 'HyV3MoeSsd', label: 'Hy-Memory 30B-A3B (Q8_0)' }, // the memory writer only (WRITER_MODELS)
17817
  'qwen35dense': { engine: 'qwen35_dense.js', exportName: 'Qwen35Dense', label: 'Underdog Saluki 27B (IQ2-mix)' },
 
19423
  const ENGINE_STORES = {
19424
  'transformers': ['transformers-cache'], 'wllama': ['wllama'], 'lfm2-webgpu': ['gguf-v1'],
19425
  'bonsai2-webgpu': ['gguf-cache-v1'], 'gemma4-webgpu': ['safetensors-cache-v1', 'bonsai-pipeline-v1'],
19426
+ 'qwen35moe-ssd': ['on disk (OPFS)'], 'minicpmv-ssd': ['on disk (OPFS)'], 'gemma4moe-ssd': ['on disk (OPFS)'], 'lfm2moe-ssd': ['on disk (OPFS)'], 'flint-ssd': ['on disk (OPFS)'], 'anyformer-glm47flash': ['on disk (OPFS)'], 'anyformer-bolt-e4b': ['transformers-cache'], 'anyformer-ling3flash': ['on disk (OPFS)'], 'anyformer-qwen36': ['on disk (OPFS)'], 'anyformer-flint': ['on disk (OPFS)'], 'anyformer-lfm25-24b': ['on disk (OPFS)'], 'anyformer-gemma26b': ['on disk (OPFS)'], 'qwen3dense': ['on disk (OPFS)'], 'qwen35dense': ['on disk (OPFS)'],
19427
  };
19428
  function parseSizeBytes(size) {
19429
  const m = /([\d.]+)\s*(GB|MB)/i.exec(String(size || ''));
 
24059
  */
24060
  const THINK_CLOSE = -2;
24061
 
24062
+ export const ANYFORMER_URL = 'https://cdn.jsdelivr.net/gh/NakliTechie/anyformer.js@v0.4.6/index.js';
24063
 
24064
  // anyformer's progress events → the engine events the SSD worker posts. The llama.cpp path reports 'download' (bytes
24065
  // into OPFS) and 'load' / 'plan'; a hand engine (diskformer back end) reports its own engine events, 'weights' with
24066
+ // bytes while it ingests and with kind 'tensors' while it uploads, and 'init'; the transformers.js back end reports
24067
+ // transformers.js's own per-file 'progress', summed here over the files seen so far (`files`, one per load) so the
24068
+ // bar does not restart at each file. Every byte event must reach the worker: the page's load watchdog ends a load
24069
+ // that reports no progress for 3 minutes (the Ling production check, 2026-10-23).
24070
+ function toEngineProgress(e, files) {
24071
  if (!e) return null;
24072
  if (e.status === 'download') return { status: 'weights', loaded: e.loaded, total: e.total };
24073
+ if (e.status === 'progress' && e.file) {
24074
+ files.set(e.file, [e.loaded || 0, e.total || 0]);
24075
+ let loaded = 0, total = 0;
24076
+ for (const [l, t] of files.values()) { loaded += l; total += t; }
24077
+ return { status: 'weights', loaded, total };
24078
+ }
24079
  if (e.status === 'weights') return { status: 'weights', kind: e.kind, loaded: e.loaded, total: e.total };
24080
  if (e.status === 'load' || e.status === 'plan' || e.status === 'init') return { status: 'init' };
24081
  return null;
24082
  }
24083
 
24084
  export class AnyformerModel {
24085
+ // A key of anyformer's MODELS.chat: anyformer's catalog then names the repo, the back end and the budget, so a
24086
+ // model moves between back ends with an anyformer release alone. repo / file / budget are for models not in it.
24087
+ static entry = null;
24088
  static repo = null;
24089
  static file = undefined; // a GGUF in the repo; undefined lets anyformer pick (the file diskformer checked)
24090
  static budget = 4e9; // GPU bytes: the llama.cpp path plans its expert slots from it; for a hand engine, the fit check
24091
  static poolBytes = undefined; // a hand engine's expert cache (LocalMind's own engines ran with 4 GiB)
24092
 
24093
  static async load(_id, opts = {}) {
24094
+ const { load, MODELS } = await import(ANYFORMER_URL);
24095
+ const entry = this.entry && MODELS.chat[this.entry], files = new Map();
24096
+ if (this.entry && !entry) throw new Error(`anyformer has no MODELS.chat.${this.entry}`);
24097
+ const session = await load(entry || this.repo, {
24098
+ ...(opts.budget || !entry ? { budget: opts.budget || this.budget } : {}), maxCtx: opts.maxCtx, signal: opts.signal, ...(this.file ? { file: this.file } : {}),
24099
  ...(this.poolBytes ? { poolBytes: this.poolBytes } : {}), ...(opts.localFile ? { localFile: opts.localFile } : {}),
24100
+ onProgress: (e) => { const ev = toEngineProgress(e, files); if (ev && opts.onProgress) opts.onProgress(ev); },
24101
  });
24102
  return new this(session);
24103
  }
 
24180
  static poolBytes = 4 * 2 ** 30;
24181
  }
24182
 
24183
+ // NVIDIA's Nemotron 3.5 Bolt E4B (dense, 4B effective, post-trained from Gemma 4 E4B), by anyformer's catalog entry:
24184
+ // at v0.4.6 a community q4f16 ONNX on transformers.js, text only, 4.9 GB, 34-39 tok/s in Chrome (2026-10-24).
24185
+ // diskformer's llama.cpp path ran it at 30 tok/s and its DFlash drafter made it slower (24.7); when a drafter pays on
24186
+ // WebGPU, anyformer moves the entry back and LocalMind changes nothing.
24187
+ export class NemotronBoltE4B extends AnyformerModel {
24188
+ static entry = 'NEMOTRON_3_5_BOLT_E4B';
24189
+ }
24190
+
24191
  // LFM2.5 24B-A2B: Liquid AI's QAD-Q4_0 GGUF (13.6 GB) on diskformer's Lfm2MoeSsd hand engine (anyformer v0.4.2+). It
24192
  // always reasons; session.think carries the markers. A 4 GiB expert cache as before.
24193
  export class Lfm25_24B extends AnyformerModel {
 
32426
  // The model loads through anyformer (convergence plan, 2026-10-22): anyformer resolves the repo to diskformer's copy of
32427
  // the LFM2.5 engine (Lfm2Mobile) and returns a session with the same generate / reset / dispose. Pinned to an exact
32428
  // version; LocalMind's anyformer_models.js pins the same one.
32429
+ const ANYFORMER_URL = 'https://cdn.jsdelivr.net/gh/NakliTechie/anyformer.js@v0.4.6/index.js';
32430
 
32431
  let anyformerLoad = null;
32432
  let model = null;