qwen-flash.json 2.2 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566
  1. {
  2. "model_code": "qwen-flash",
  3. "model_intro": "Qwen3系列Flash模型,实现思考模式和非思考模式的有效融合,可在对话中切换模式。复杂推理类任务性能优秀,指令遵循、文本理解等能力显著提高。支持1M上下文长度,按照上下文长度进行阶梯计费。",
  4. "model_tags": ["Qwen3", "深度思考", "文本生成"],
  5. "model_capabilities": {
  6. "input_modalities": ["text"],
  7. "output_modalities": ["text"],
  8. "features": {
  9. "model_experience": true,
  10. "function_calling": true,
  11. "structured_output": true,
  12. "web_search": true,
  13. "prefix_continuation": true,
  14. "cache": true,
  15. "batch_inference": true,
  16. "model_tuning": false
  17. }
  18. },
  19. "model_pricing": [
  20. {
  21. "input_range": "输入<=128K",
  22. "unit": "元/千tokens",
  23. "input": 0.00015,
  24. "input_cache_hit": 0.00003,
  25. "input_batch": 0.000075,
  26. "output": 0.0015,
  27. "output_batch": 0.00075,
  28. "explicit_cache_create": 0.0001875,
  29. "explicit_cache_hit": 0.000015
  30. },
  31. {
  32. "input_range": "128K<输入<=256K",
  33. "unit": "元/千tokens",
  34. "input": 0.0006,
  35. "input_cache_hit": 0.00012,
  36. "input_batch": 0.0003,
  37. "output": 0.006,
  38. "output_batch": 0.003,
  39. "explicit_cache_create": 0.00075,
  40. "explicit_cache_hit": 0.00006
  41. },
  42. {
  43. "input_range": "256K<输入<=1M",
  44. "unit": "元/千tokens",
  45. "input": 0.0012,
  46. "input_cache_hit": 0.00024,
  47. "input_batch": 0.0006,
  48. "output": 0.012,
  49. "output_batch": 0.006,
  50. "explicit_cache_create": 0.0015,
  51. "explicit_cache_hit": 0.00012
  52. }
  53. ],
  54. "model_limits": {
  55. "max_input_length": "997K",
  56. "max_output_length": "32K",
  57. "rpm": 15000,
  58. "tpm": 10000000,
  59. "context_length": "1M"
  60. },
  61. "api_examples": {
  62. "curl": [
  63. "curl -X POST https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions \\\n-H \"Authorization: Bearer $DASHSCOPE_API_KEY\" \\\n-H \"Content-Type: application/json\" \\\n-d '{\n \"model\": \"qwen-flash\",\n \"messages\": [\n {\n \"role\": \"user\", \n \"content\": \"你是谁\"\n }\n ],\n \"stream\": true,\n \"enable_thinking\": true\n}'"
  64. ]
  65. }
  66. }