nsarrazin commited on
Commit
bb8ac6e
·
1 Parent(s): 9111ea7

feat(huggingchat): update available models

Browse files

- remove llama 3.2
- remove qwen coder
- add qwen 2.5 VL 32B

Files changed (1) hide show
  1. chart/env/prod.yaml +42 -74
chart/env/prod.yaml CHANGED
@@ -230,40 +230,6 @@ envVars:
230
  }
231
  ]
232
  },
233
- {
234
- "name": "Qwen/Qwen2.5-Coder-32B-Instruct",
235
- "description": "Qwen's latest coding model, in its biggest size yet. SOTA on many coding benchmarks.",
236
- "modelUrl": "https://huggingface.co/Qwen/Qwen2.5-Coder-32B-Instruct",
237
- "websiteUrl": "https://qwenlm.github.io/blog/qwen2.5-coder-family/",
238
- "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/qwen-logo.png",
239
- "preprompt": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant.",
240
- "parameters": {
241
- "stop": ["<|im_end|>", "<|endoftext|>"],
242
- "temperature": 0.6,
243
- "truncate": 28672,
244
- "max_new_tokens": 3072
245
- },
246
- "promptExamples": [
247
- {
248
- "title": "Build a webapp",
249
- "prompt": "Create a simple to-do list application where users can:\n- Add new tasks.\n- Mark tasks as complete.\n- Delete completed tasks.\nThe tasks should persist in the browser's local storage so that they remain available even after a page reload.\n"
250
- },
251
- {
252
- "title": "Create a REST API",
253
- "prompt": "Build a simple REST API using Node.js, TypeScript and Express:\n- POST /items: Accepts a JSON body with name and quantity and adds a new item.\n- GET /items: Returns a list of all items.\n- PUT /items/:id: Updates the name or quantity of an item by its id.\n- DELETE /items/:id: Removes an item by its id.\nUse an in-memory array as the data store (no need for a database). Include basic error handling (e.g., item not found)."
254
- },
255
- {
256
- "title": "Simple website",
257
- "prompt": "Generate a snazzy static landing page for a local coffee shop using HTML and CSS. You can use tailwind using <script src='https://cdn.tailwindcss.com'></script>."
258
- }
259
- ],
260
- "endpoints": [
261
- {
262
- "type": "openai",
263
- "baseURL": "https://internal.api-inference.huggingface.co/models/Qwen/Qwen2.5-Coder-32B-Instruct/v1"
264
- }
265
- ]
266
- },
267
  {
268
  "name": "google/gemma-3-27b-it",
269
  "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/google-logo.png",
@@ -331,17 +297,11 @@ envVars:
331
  ]
332
  },
333
  {
334
- "name": "meta-llama/Llama-3.2-11B-Vision-Instruct",
335
- "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/meta-logo.png",
336
- "description": "The latest multimodal model from Meta! Supports image inputs natively.",
337
- "websiteUrl": "https://llama.com/",
338
  "multimodal": true,
339
- "parameters": {
340
- "stop": ["<|eot_id|>", "<|im_end|>"],
341
- "temperature": 0.6,
342
- "truncate": 14336,
343
- "max_new_tokens": 1536
344
- },
345
  "promptExamples": [
346
  {
347
  "title": "Write an email",
@@ -359,12 +319,12 @@ envVars:
359
  "endpoints": [
360
  {
361
  "type": "openai",
362
- "baseURL": "https://internal.api-inference.huggingface.co/models/meta-llama/Llama-3.2-11B-Vision-Instruct/v1",
363
  "multimodal": {
364
  "image": {
365
  "maxSizeInMB": 10,
366
- "maxWidth": 560,
367
- "maxHeight": 560,
368
  "supportedMimeTypes": ["image/png", "image/jpeg", "image/webp"],
369
  "preferredMimeType": "image/webp"
370
  }
@@ -372,33 +332,6 @@ envVars:
372
  }
373
  ]
374
  },
375
- {
376
- "name": "NousResearch/Hermes-3-Llama-3.1-8B",
377
- "description": "Nous Research's latest Hermes 3 release in 8B size. Follows instruction closely.",
378
- "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/nous-logo.png",
379
- "websiteUrl": "https://nousresearch.com/",
380
- "modelUrl": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B",
381
- "promptExamples": [
382
- {
383
- "title": "Write an email",
384
- "prompt": "As a restaurant owner, write a professional email to the supplier to get these products every week: \n\n- Wine (x10)\n- Eggs (x24)\n- Bread (x12)"
385
- },
386
- {
387
- "title": "Code a game",
388
- "prompt": "Code a basic snake game in python, give explanations for each step."
389
- },
390
- {
391
- "title": "Recipe help",
392
- "prompt": "How do I make a delicious lemon cheesecake?"
393
- }
394
- ],
395
- "parameters": {
396
- "stop": ["<|im_end|>"],
397
- "temperature": 0.6,
398
- "truncate": 14336,
399
- "max_new_tokens": 1536
400
- }
401
- },
402
  {
403
  "name": "microsoft/Phi-4",
404
  "description": "One of the best small models, super fast for simple tasks.",
@@ -433,6 +366,33 @@ envVars:
433
  }
434
  ]
435
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
436
  {
437
  "name": "internal/task",
438
  "tokenizer" : "NousResearch/Hermes-3-Llama-3.1-8B",
@@ -507,6 +467,14 @@ envVars:
507
  {
508
  "name": "microsoft/Phi-3.5-mini-instruct",
509
  "transferTo": "microsoft/Phi-4"
 
 
 
 
 
 
 
 
510
  }
511
  ]
512
  PUBLIC_ORIGIN: "https://huggingface.co"
 
230
  }
231
  ]
232
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
233
  {
234
  "name": "google/gemma-3-27b-it",
235
  "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/google-logo.png",
 
297
  ]
298
  },
299
  {
300
+ "name": "Qwen/Qwen2.5-VL-32B-Instruct",
301
+ "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/qwen-logo.png",
302
+ "description": "The latest multimodal model from Qwen! Supports image inputs natively.",
303
+ "websiteUrl": "https://qwenlm.github.io/blog/qwen2.5-vl/",
304
  "multimodal": true,
 
 
 
 
 
 
305
  "promptExamples": [
306
  {
307
  "title": "Write an email",
 
319
  "endpoints": [
320
  {
321
  "type": "openai",
322
+ "baseURL": "https://lf91qeosuambouj4.us-east-1.aws.endpoints.huggingface.cloud/v1",
323
  "multimodal": {
324
  "image": {
325
  "maxSizeInMB": 10,
326
+ "maxWidth": 1024,
327
+ "maxHeight": 1024,
328
  "supportedMimeTypes": ["image/png", "image/jpeg", "image/webp"],
329
  "preferredMimeType": "image/webp"
330
  }
 
332
  }
333
  ]
334
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
335
  {
336
  "name": "microsoft/Phi-4",
337
  "description": "One of the best small models, super fast for simple tasks.",
 
366
  }
367
  ]
368
  },
369
+ {
370
+ "name": "NousResearch/Hermes-3-Llama-3.1-8B",
371
+ "description": "Nous Research's latest Hermes 3 release in 8B size. Follows instruction closely.",
372
+ "logoUrl": "https://huggingface.co/datasets/huggingchat/models-logo/resolve/main/nous-logo.png",
373
+ "websiteUrl": "https://nousresearch.com/",
374
+ "modelUrl": "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-8B",
375
+ "promptExamples": [
376
+ {
377
+ "title": "Write an email",
378
+ "prompt": "As a restaurant owner, write a professional email to the supplier to get these products every week: \n\n- Wine (x10)\n- Eggs (x24)\n- Bread (x12)"
379
+ },
380
+ {
381
+ "title": "Code a game",
382
+ "prompt": "Code a basic snake game in python, give explanations for each step."
383
+ },
384
+ {
385
+ "title": "Recipe help",
386
+ "prompt": "How do I make a delicious lemon cheesecake?"
387
+ }
388
+ ],
389
+ "parameters": {
390
+ "stop": ["<|im_end|>"],
391
+ "temperature": 0.6,
392
+ "truncate": 14336,
393
+ "max_new_tokens": 1536
394
+ }
395
+ },
396
  {
397
  "name": "internal/task",
398
  "tokenizer" : "NousResearch/Hermes-3-Llama-3.1-8B",
 
467
  {
468
  "name": "microsoft/Phi-3.5-mini-instruct",
469
  "transferTo": "microsoft/Phi-4"
470
+ },
471
+ {
472
+ "name": "Qwen/Qwen2.5-Coder-32B-Instruct",
473
+ "transferTo": "Qwen/QwQ-32B"
474
+ },
475
+ {
476
+ "name": "meta-llama/Llama-3.2-11B-Vision-Instruct",
477
+ "transferTo" : "Qwen/Qwen2.5-VL-32B-Instruct"
478
  }
479
  ]
480
  PUBLIC_ORIGIN: "https://huggingface.co"