akhaliq HF Staff commited on
Commit
9e44e59
·
1 Parent(s): 0e4849d

remove qwen models

Browse files
Files changed (1) hide show
  1. app.py +2 -98
app.py CHANGED
@@ -2281,11 +2281,6 @@ AVAILABLE_MODELS = [
2281
  "id": "MiniMaxAI/MiniMax-M1-80k",
2282
  "description": "MiniMax M1 model for code generation and general tasks"
2283
  },
2284
- {
2285
- "name": "Qwen3-235B-A22B",
2286
- "id": "Qwen/Qwen3-235B-A22B",
2287
- "description": "Qwen3-235B-A22B model for code generation and general tasks"
2288
- },
2289
  {
2290
  "name": "GLM-4.5",
2291
  "id": "zai-org/GLM-4.5",
@@ -2301,66 +2296,6 @@ AVAILABLE_MODELS = [
2301
  "id": "zai-org/GLM-4.6",
2302
  "description": "GLM-4.6 model for advanced code generation and general tasks"
2303
  },
2304
- {
2305
- "name": "Qwen3-235B-A22B-Instruct-2507",
2306
- "id": "Qwen/Qwen3-235B-A22B-Instruct-2507",
2307
- "description": "Qwen3-235B-A22B-Instruct-2507 model for code generation and general tasks"
2308
- },
2309
- {
2310
- "name": "Qwen3-Coder-480B-A35B-Instruct",
2311
- "id": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
2312
- "description": "Qwen3-Coder-480B-A35B-Instruct model for advanced code generation and programming tasks"
2313
- },
2314
- {
2315
- "name": "Qwen3-32B",
2316
- "id": "Qwen/Qwen3-32B",
2317
- "description": "Qwen3-32B model for code generation and general tasks"
2318
- },
2319
- {
2320
- "name": "Qwen3-4B-Instruct-2507",
2321
- "id": "Qwen/Qwen3-4B-Instruct-2507",
2322
- "description": "Qwen3-4B-Instruct-2507 model for code generation and general tasks"
2323
- },
2324
- {
2325
- "name": "Qwen3-4B-Thinking-2507",
2326
- "id": "Qwen/Qwen3-4B-Thinking-2507",
2327
- "description": "Qwen3-4B-Thinking-2507 model with advanced reasoning capabilities for code generation and general tasks"
2328
- },
2329
- {
2330
- "name": "Qwen3-235B-A22B-Thinking",
2331
- "id": "Qwen/Qwen3-235B-A22B-Thinking-2507",
2332
- "description": "Qwen3-235B-A22B-Thinking model with advanced reasoning capabilities"
2333
- },
2334
- {
2335
- "name": "Qwen3-Next-80B-A3B-Thinking",
2336
- "id": "Qwen/Qwen3-Next-80B-A3B-Thinking",
2337
- "description": "Qwen3-Next-80B-A3B-Thinking model with advanced reasoning capabilities via Hyperbolic"
2338
- },
2339
- {
2340
- "name": "Qwen3-Next-80B-A3B-Instruct",
2341
- "id": "Qwen/Qwen3-Next-80B-A3B-Instruct",
2342
- "description": "Qwen3-Next-80B-A3B-Instruct model for code generation and general tasks via Hyperbolic"
2343
- },
2344
- {
2345
- "name": "Qwen3-30B-A3B-Instruct-2507",
2346
- "id": "qwen3-30b-a3b-instruct-2507",
2347
- "description": "Qwen3-30B-A3B-Instruct model via Alibaba Cloud DashScope API"
2348
- },
2349
- {
2350
- "name": "Qwen3-30B-A3B-Thinking-2507",
2351
- "id": "qwen3-30b-a3b-thinking-2507",
2352
- "description": "Qwen3-30B-A3B-Thinking model with advanced reasoning via Alibaba Cloud DashScope API"
2353
- },
2354
- {
2355
- "name": "Qwen3-Coder-30B-A3B-Instruct",
2356
- "id": "qwen3-coder-30b-a3b-instruct",
2357
- "description": "Qwen3-Coder-30B-A3B-Instruct model for advanced code generation via Alibaba Cloud DashScope API"
2358
- },
2359
- {
2360
- "name": "Qwen3-Coder-Plus-2025-09-23",
2361
- "id": "qwen3-coder-plus-2025-09-23",
2362
- "description": "Qwen3-Coder-Plus-2025-09-23 model - latest advanced code generation model via Alibaba Cloud DashScope API"
2363
- },
2364
  {
2365
  "name": "Cohere Command-A Reasoning 08-2025",
2366
  "id": "CohereLabs/command-a-reasoning-08-2025",
@@ -2453,11 +2388,6 @@ AVAILABLE_MODELS = [
2453
  "name": "Qwen3 Max Preview",
2454
  "id": "qwen3-max-preview",
2455
  "description": "Qwen3 Max Preview model via DashScope International API"
2456
- },
2457
- {
2458
- "name": "Qwen3-Max-2025-09-23",
2459
- "id": "qwen3-max-2025-09-23",
2460
- "description": "Qwen3-Max-2025-09-23 model - latest flagship model via Alibaba Cloud DashScope API"
2461
  }
2462
  ]
2463
 
@@ -2497,12 +2427,6 @@ def get_inference_client(model_id, provider="auto"):
2497
  api_key=os.getenv("DASHSCOPE_API_KEY"),
2498
  base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2499
  )
2500
- elif model_id == "qwen3-coder-plus-2025-09-23":
2501
- # Use DashScope OpenAI client for Qwen3-Coder-Plus-2025-09-23 model
2502
- return OpenAI(
2503
- api_key=os.getenv("DASHSCOPE_API_KEY"),
2504
- base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2505
- )
2506
  elif model_id == "gpt-5":
2507
  # Use Poe (OpenAI-compatible) client for GPT-5 model
2508
  return OpenAI(
@@ -2545,12 +2469,6 @@ def get_inference_client(model_id, provider="auto"):
2545
  api_key=os.getenv("DASHSCOPE_API_KEY"),
2546
  base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2547
  )
2548
- elif model_id == "qwen3-max-2025-09-23":
2549
- # Use DashScope OpenAI client for Qwen3-Max-2025-09-23 model
2550
- return OpenAI(
2551
- api_key=os.getenv("DASHSCOPE_API_KEY"),
2552
- base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2553
- )
2554
  elif model_id == "openrouter/sonoma-dusk-alpha":
2555
  # Use OpenRouter client for Sonoma Dusk Alpha model
2556
  return OpenAI(
@@ -2632,20 +2550,6 @@ def get_inference_client(model_id, provider="auto"):
2632
  provider = "groq"
2633
  elif model_id == "moonshotai/Kimi-K2-Instruct":
2634
  provider = "groq"
2635
- elif model_id == "Qwen/Qwen3-235B-A22B":
2636
- provider = "cerebras"
2637
- elif model_id == "Qwen/Qwen3-235B-A22B-Instruct-2507":
2638
- provider = "cerebras"
2639
- elif model_id == "Qwen/Qwen3-32B":
2640
- provider = "cerebras"
2641
- elif model_id == "Qwen/Qwen3-235B-A22B-Thinking-2507":
2642
- provider = "cerebras"
2643
- elif model_id == "Qwen/Qwen3-Coder-480B-A35B-Instruct":
2644
- provider = "cerebras"
2645
- elif model_id == "Qwen/Qwen3-Next-80B-A3B-Thinking":
2646
- provider = "hyperbolic"
2647
- elif model_id == "Qwen/Qwen3-Next-80B-A3B-Instruct":
2648
- provider = "novita"
2649
  elif model_id == "deepseek-ai/DeepSeek-V3.1":
2650
  provider = "novita"
2651
  elif model_id == "deepseek-ai/DeepSeek-V3.1-Terminus":
@@ -7241,7 +7145,7 @@ def generate_requirements_txt_with_llm(import_statements):
7241
 
7242
  # Use a lightweight model for this task
7243
  try:
7244
- client = get_inference_client("Qwen/Qwen3-Coder-480B-A35B-Instruct", "auto")
7245
 
7246
  imports_text = '\n'.join(import_statements)
7247
 
@@ -7288,7 +7192,7 @@ Generate a comprehensive requirements.txt that ensures the application will work
7288
  ]
7289
 
7290
  response = client.chat.completions.create(
7291
- model="Qwen/Qwen3-Coder-480B-A35B-Instruct",
7292
  messages=messages,
7293
  max_tokens=1024,
7294
  temperature=0.1
 
2281
  "id": "MiniMaxAI/MiniMax-M1-80k",
2282
  "description": "MiniMax M1 model for code generation and general tasks"
2283
  },
 
 
 
 
 
2284
  {
2285
  "name": "GLM-4.5",
2286
  "id": "zai-org/GLM-4.5",
 
2296
  "id": "zai-org/GLM-4.6",
2297
  "description": "GLM-4.6 model for advanced code generation and general tasks"
2298
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2299
  {
2300
  "name": "Cohere Command-A Reasoning 08-2025",
2301
  "id": "CohereLabs/command-a-reasoning-08-2025",
 
2388
  "name": "Qwen3 Max Preview",
2389
  "id": "qwen3-max-preview",
2390
  "description": "Qwen3 Max Preview model via DashScope International API"
 
 
 
 
 
2391
  }
2392
  ]
2393
 
 
2427
  api_key=os.getenv("DASHSCOPE_API_KEY"),
2428
  base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2429
  )
 
 
 
 
 
 
2430
  elif model_id == "gpt-5":
2431
  # Use Poe (OpenAI-compatible) client for GPT-5 model
2432
  return OpenAI(
 
2469
  api_key=os.getenv("DASHSCOPE_API_KEY"),
2470
  base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",
2471
  )
 
 
 
 
 
 
2472
  elif model_id == "openrouter/sonoma-dusk-alpha":
2473
  # Use OpenRouter client for Sonoma Dusk Alpha model
2474
  return OpenAI(
 
2550
  provider = "groq"
2551
  elif model_id == "moonshotai/Kimi-K2-Instruct":
2552
  provider = "groq"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2553
  elif model_id == "deepseek-ai/DeepSeek-V3.1":
2554
  provider = "novita"
2555
  elif model_id == "deepseek-ai/DeepSeek-V3.1-Terminus":
 
7145
 
7146
  # Use a lightweight model for this task
7147
  try:
7148
+ client = get_inference_client("zai-org/GLM-4.6", "auto")
7149
 
7150
  imports_text = '\n'.join(import_statements)
7151
 
 
7192
  ]
7193
 
7194
  response = client.chat.completions.create(
7195
+ model="zai-org/GLM-4.6",
7196
  messages=messages,
7197
  max_tokens=1024,
7198
  temperature=0.1