New request
#2926: 需要对token用量的精确统计，包括输入输出，缓存命中之类的
We'll provision a sandbox, run an agent against the issue, and open a draft PR. You can pull the branch and iterate from there.
Issue
enhancement
No issue body. The agent will reason from the title and the repo's architecture context alone.
Assessmentadvisory
feature●● medium70% confidence
Exposes LLM token metrics (input/output/cache) in the UI, requiring propagation of provider telemetry through the API layer and new frontend components.
Likely files
- api/routes.py
- api/models.py
- static/index.html
- static/app.js
- static/panels.js
Create the request