Skip to content

docs: plan local model benchmarking and routing - #97

Open
XYAIStudio wants to merge 2 commits into
mainfrom
codex/local-model-auto-routing-plan
Open

XYAIStudio wants to merge 2 commits into
mainfrom
codex/local-model-auto-routing-plan

Conversation

@XYAIStudio

Copy link
Copy Markdown
Owner

问题与结果

普通用户难以判断本地模型是否适合实时对话、长时任务、工具调用或视觉任务,单看模型大小或一次生成速度也会产生误导。本 PR 提供一份可直接实施的系统规划:先验证能力硬约束,再按场景完成轻量基准、评分、推荐、线程级稳定路由和故障降级。

规划内容

  • 两级基准:首次轻量测速与后台完整复测
  • 指标体系:首字时间、生成速度、上下文、稳定性、资源占用及工具/视觉/JSON 能力
  • 五类场景评分和推荐置信度
  • 线程粘性、迟滞阈值、故障分类与回退链
  • 数据表、API、前端交互、隐私和可观测性设计
  • M1–M4 分阶段交付、验收标准、测试矩阵和成功指标

验证

  • make all(pre-commit):通过
  • Dashboard production build:通过
  • 文档提交未包含工作区中原有的未提交代码改动

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant