config-center/outputs/field-audit/official-review-checklist.md

Official Review Checklist

GeneratedAt: 2026-04-25T16:15:04.854Z Source: /Users/xieyuanxiang/config-center/outputs/field-audit/zenmux-sync-report.json

Purpose: verify unresolved models with official vendor docs before write-back.

Summary

• actual models: 160
• unresolved models: 103
• providers involved: 27
• null sampling fields: 0

Provider: baichuan

• official docs:
  • https://platform.baichuan-ai.com/docs
• pending models:
  • Baichuan-M3-Plus (compute/providers/baichuan.json) -> no ZenMux match
  • Baichuan-M3 (compute/providers/baichuan.json) -> no ZenMux match
  • Baichuan-M2-Plus (compute/providers/baichuan.json) -> no ZenMux match
  • Baichuan-M2 (compute/providers/baichuan.json) -> no ZenMux match

Provider: baidu

• official docs:
  • https://cloud.baidu.com/doc/qianfan/
• pending models:
  • ernie-4.5-turbo-128k (compute/providers/baidu.json) -> no ZenMux match
  • ernie-4.5-turbo-20260402 (compute/providers/baidu.json) -> no ZenMux match

Provider: cohere

• official docs:
  • https://docs.cohere.com/docs/models
  • https://cohere.com/pricing
• pending models:
  • command-a-03-2025 (compute/providers/cohere.json) -> no ZenMux match
  • command-r7b-12-2024 (compute/providers/cohere.json) -> no ZenMux match
  • embed-v4.0 (compute/providers/cohere.json) -> no ZenMux match
  • rerank-v3.5 (compute/providers/cohere.json) -> no ZenMux match

Provider: dashscope

• official docs:
  • https://help.aliyun.com/zh/model-studio/getting-started/models
  • https://help.aliyun.com/zh/model-studio/pricing
• pending models:
  • qwen3.6-flash (compute/providers/dashscope.json) -> ambiguous match candidates: qwen/qwen3.6-plus,qwen/qwen3.5-flash
  • qwen-max (compute/providers/dashscope.json) -> no ZenMux match
  • qwen-plus (compute/providers/dashscope.json) -> no ZenMux match
  • qwen-turbo (compute/providers/dashscope.json) -> no ZenMux match
  • qwen-long (compute/providers/dashscope.json) -> no ZenMux match
  • qwen3-vl-flash (compute/providers/dashscope.json) -> ambiguous match candidates: qwen/qwen3.5-flash,qwen/qwen3-vl-plus
  • text-embedding-v3 (compute/providers/dashscope.json) -> no ZenMux match
  • text-embedding-v4 (compute/providers/dashscope.json) -> no ZenMux match
  • qwen3-rerank (compute/providers/dashscope.json) -> no ZenMux match
  • cosyvoice-v2 (compute/providers/dashscope.json) -> no ZenMux match
  • paraformer-v2 (compute/providers/dashscope.json) -> no ZenMux match
  • wanx-v2 (compute/providers/dashscope.json) -> no ZenMux match
  • wanx-video (compute/providers/dashscope.json) -> no ZenMux match
  • cosyvoice-clone (compute/providers/dashscope.json) -> no ZenMux match
  • qwen-omni-turbo (compute/providers/dashscope.json) -> no ZenMux match

Provider: google

• official docs:
  • https://ai.google.dev/gemini-api/docs/models
  • https://ai.google.dev/pricing
• pending models:
  • text-embedding-005 (compute/providers/google.json) -> no ZenMux match

Provider: infini

• official docs:
  • (add link manually)
• pending models:
  • deepseek-v3 (compute/coding-plans/infini-coding.json) -> no ZenMux match

Provider: internal-testing

• official docs:
  • (add link manually)
• pending models:
  • MiniMax-M2.7-highspeed (compute/providers/internal-testing.json) -> no ZenMux match
  • glm-5.1 (compute/providers/internal-testing.json) -> no ZenMux match
  • glm-5 (compute/providers/internal-testing.json) -> no ZenMux match
  • glm-5-turbo (compute/providers/internal-testing.json) -> no ZenMux match
  • glm-4.7 (compute/providers/internal-testing.json) -> no ZenMux match
  • kimi-k2.6-code-preview (compute/providers/internal-testing.json) -> no ZenMux match
  • kimi-k2.5 (compute/providers/internal-testing.json) -> no ZenMux match
  • MiniMax-M2.5 (compute/providers/internal-testing.json) -> no ZenMux match
  • qwen3.6-plus (compute/providers/internal-testing.json) -> no ZenMux match
  • qwen3.5-plus (compute/providers/internal-testing.json) -> no ZenMux match
  • qwen3.5-35b-a3b (compute/providers/internal-testing.json) -> no ZenMux match
  • qwen3.5-27b (compute/providers/internal-testing.json) -> no ZenMux match
  • qwen3-max-2026-01-23 (compute/providers/internal-testing.json) -> no ZenMux match
  • doubao-seed-2-0-code-preview-260215 (compute/providers/internal-testing.json) -> no ZenMux match

Provider: kling

• official docs:
  • (add link manually)
• pending models:
  • kling-v2-5-turbo (compute/providers/kling.json) -> no ZenMux match
  • kling-v2-5-turbo-pro (compute/providers/kling.json) -> no ZenMux match
  • kling-v2 (compute/providers/kling.json) -> no ZenMux match
  • kling-v2-master (compute/providers/kling.json) -> no ZenMux match

Provider: kwai

• official docs:
  • (add link manually)
• pending models:
  • kwai-coder (compute/coding-plans/kwai-coding.json) -> no ZenMux match

Provider: lingyiwanwu

• official docs:
  • (add link manually)
• pending models:
  • yi-lightning (compute/providers/lingyiwanwu.json) -> no ZenMux match
  • yi-vision-v2 (compute/providers/lingyiwanwu.json) -> no ZenMux match

Provider: local-whisper

• official docs:
  • (add link manually)
• pending models:
  • whisper-large-v3 (compute/providers/local-whisper.json) -> no ZenMux match

Provider: minimax

• official docs:
  • https://platform.minimax.io/docs/api-reference/api-overview
  • https://platform.minimax.io/docs/guides/pricing-paygo
• pending models:
  • MiniMax-M1 (compute/coding-plans/minimax-coding.json) -> no ZenMux match
  • MiniMax-M2.5-highspeed (compute/providers/minimax.json) -> ambiguous match candidates: minimax/minimax-m2.5,minimax/minimax-m2.7-highspeed,minimax/minimax-m2.5-lightning,minimax/minimax-m2,minimax/minimax-m2.7
  • MiniMax-M2.1-highspeed (compute/providers/minimax.json) -> ambiguous match candidates: minimax/minimax-m2.1,minimax/minimax-m2.7-highspeed,minimax/minimax-m2,minimax/minimax-m2.7,minimax/minimax-m2.5
  • MiniMax-Text-01 (compute/providers/minimax.json) -> no ZenMux match

Provider: mistral

• official docs:
  • https://docs.mistral.ai/getting-started/models
  • https://mistral.ai/pricing
• pending models:
  • mistral-small-latest (compute/providers/mistral.json) -> no ZenMux match
  • codestral-latest (compute/providers/mistral.json) -> no ZenMux match

Provider: moonshot

• official docs:
  • https://platform.moonshot.cn/docs/guide/overview
  • https://platform.moonshot.cn/docs/pricing/chat
• pending models:
  • kimi-k2 (compute/coding-plans/moonshot-coding.json) -> ambiguous match candidates: moonshotai/kimi-k2.6,moonshotai/kimi-k2.5,moonshotai/kimi-k2-thinking,moonshotai/kimi-k2-0905,moonshotai/kimi-k2-0711
  • kimi-k2 (compute/providers/moonshot.json) -> ambiguous match candidates: moonshotai/kimi-k2.6,moonshotai/kimi-k2.5,moonshotai/kimi-k2-thinking,moonshotai/kimi-k2-0905,moonshotai/kimi-k2-0711
  • moonshot-v1-8k (compute/providers/moonshot.json) -> no ZenMux match
  • moonshot-v1-32k (compute/providers/moonshot.json) -> no ZenMux match
  • moonshot-v1-128k (compute/providers/moonshot.json) -> no ZenMux match

Provider: moorethread

• official docs:
  • (add link manually)
• pending models:
  • mt-coder (compute/coding-plans/moorethread-coding.json) -> no ZenMux match

Provider: ollama

• official docs:
  • (add link manually)
• pending models:
  • llama3.1:70b (compute/providers/ollama.json) -> no ZenMux match

Provider: openai

• official docs:
  • https://platform.openai.com/docs/models
  • https://platform.openai.com/docs/pricing
• pending models:
  • text-embedding-3-small (compute/providers/openai.json) -> no ZenMux match
  • text-embedding-3-large (compute/providers/openai.json) -> no ZenMux match
  • tts-1 (compute/providers/openai.json) -> no ZenMux match
  • tts-1-hd (compute/providers/openai.json) -> no ZenMux match
  • whisper-1 (compute/providers/openai.json) -> no ZenMux match
  • o3 (compute/providers/openai.json) -> no ZenMux match
  • o3-pro (compute/providers/openai.json) -> no ZenMux match
  • o3-mini (compute/providers/openai.json) -> no ZenMux match
  • dall-e-3 (compute/providers/openai.json) -> no ZenMux match
  • gpt-4o-realtime (compute/providers/openai.json) -> ambiguous match candidates: openai/gpt-4o,openai/gpt-4o-mini
  • gpt-4o-realtime-preview (compute/providers/openai.json) -> ambiguous match candidates: openai/gpt-4o,openai/gpt-4o-mini

Provider: openrouter

• official docs:
  • https://openrouter.ai/models
• pending models:
  • openrouter/auto (compute/providers/openrouter.json) -> no ZenMux match
  • openai/gpt-oss-120b:free (compute/providers/openrouter.json) -> no ZenMux match
  • qwen/qwen3-coder:free (compute/providers/openrouter.json) -> no ZenMux match

Provider: perplexity

• official docs:
  • https://docs.perplexity.ai
• pending models:
  • sonar-pro (compute/providers/perplexity.json) -> no ZenMux match
  • sonar-reasoning-pro (compute/providers/perplexity.json) -> no ZenMux match
  • sonar (compute/providers/perplexity.json) -> no ZenMux match

Provider: siliconflow

• official docs:
  • https://www.siliconflow.com/models
  • https://siliconflow.cn/pricing
• pending models:
  • Qwen/Qwen3-Coder-480B-A35B-Instruct (compute/providers/siliconflow.json) -> no ZenMux match
  • Qwen/Qwen3-235B-A22B-Instruct-2507 (compute/providers/siliconflow.json) -> ambiguous match candidates: qwen/qwen3-235b-a22b-2507,qwen/qwen3-235b-a22b-thinking-2507
  • BAAI/bge-m3 (compute/providers/siliconflow.json) -> no ZenMux match

Provider: stability

• official docs:
  • (add link manually)
• pending models:
  • stable-diffusion-3.5-large (compute/providers/stability.json) -> no ZenMux match

Provider: tencent

• official docs:
  • https://cloud.tencent.com/document/product/1729
• pending models:
  • hunyuan-turbos-latest (compute/providers/tencent.json) -> no ZenMux match

Provider: volcengine

• official docs:
  • https://www.volcengine.com/docs/82379
• pending models:
  • ark-code-latest (compute/coding-plans/volcengine-coding.json) -> no ZenMux match
  • doubao-2.0-pro (compute/providers/volcengine.json) -> ambiguous match candidates: bytedance/doubao-seed-2.0-pro,bytedance/doubao-seed-2.0-mini,bytedance/doubao-seed-2.0-lite,bytedance/doubao-seed-2.0-code
  • doubao-seed-1.6 (compute/providers/volcengine.json) -> ambiguous match candidates: bytedance/doubao-seed-1.8,bytedance/doubao-seed-code
  • doubao-seed-1.6-lite (compute/providers/volcengine.json) -> ambiguous match candidates: bytedance/doubao-seed-1.8,bytedance/doubao-seed-2.0-lite
  • deepseek-v3.2 (compute/providers/volcengine.json) -> no ZenMux match
  • deepseek-r1 (compute/providers/volcengine.json) -> no ZenMux match
  • kimi-k2-volcengine (compute/providers/volcengine.json) -> no ZenMux match
  • glm-4-7 (compute/providers/volcengine.json) -> no ZenMux match
  • doubao-embedding (compute/providers/volcengine.json) -> no ZenMux match
  • volc-mega-tts-clone (compute/providers/volcengine.json) -> no ZenMux match
  • volc-realtime-voice (compute/providers/volcengine.json) -> no ZenMux match
  • volc-simultaneous (compute/providers/volcengine.json) -> no ZenMux match
  • volc-translation (compute/providers/volcengine.json) -> no ZenMux match

Provider: xai

• official docs:
  • https://docs.x.ai/docs/models
• pending models:
  • grok-4.20-0309-reasoning (compute/providers/xai.json) -> ambiguous match candidates: x-ai/grok-4-fast-non-reasoning,x-ai/grok-4,x-ai/grok-4.2-fast-non-reasoning,x-ai/grok-4.1-fast-non-reasoning
  • grok-4-1-fast-reasoning (compute/providers/xai.json) -> ambiguous match candidates: x-ai/grok-4.1-fast-non-reasoning,x-ai/grok-4.1-fast,x-ai/grok-4-fast-non-reasoning,x-ai/grok-4-fast,x-ai/grok-4.2-fast-non-reasoning

Provider: xunfei

• official docs:
  • https://www.xfyun.cn/doc/
• pending models:
  • spark-x (compute/providers/xunfei.json) -> no ZenMux match
  • 4.0Ultra (compute/providers/xunfei.json) -> no ZenMux match

Provider: zhipu

• official docs:
  • https://docs.bigmodel.cn/cn/guide/models/text/
  • https://www.bigmodel.cn/pricing
• pending models:
  • glm-4.7-thinking (compute/providers/zhipu.json) -> ambiguous match candidates: z-ai/glm-4.7,z-ai/glm-4.7-flashx,z-ai/glm-4.7-flash-free,z-ai/glm-4.6v,z-ai/glm-4.6

Provider: zhipu-embedding

• official docs:
  • https://docs.bigmodel.cn/cn/guide/models/embedding
• pending models:
  • embedding-3 (compute/providers/zhipu-embedding.json) -> no ZenMux match

Field-by-Field Verification Rule

• modelName: must exactly match provider API model ID or official alias rule.
• contextWindow: use official model spec limit.
• maxOutputTokens: use official output cap; do not infer from context.
• inputPrice/outputPrice: use official published API pricing and the provider currency.
• capabilities: project taxonomy field; keep local semantics unless official docs clearly contradict the capability.
• defaultTemperature/defaultTopP: only fill with numeric values when provider docs define a safe default; omit unsupported fields instead of using null.