MoonshotAI / MoonshotAI/kimi-cli
报错信息也消耗了token || The error message also consumes tokens
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 11.4k
- Forks
- 1.3k
- Avg merge
- 9h 47m
- Merged PRs (30d)
- 2
Description
What version of Kimi Code CLI is running?
1.37.0
Which open platform/subscription were you using?
kimi code plan 99
Which model were you using?
kimi 2.6
What platform is your computer?
No response
What issue are you seeing?
token 消耗的太快，好多次报错了，但是token也消耗了。目前的计算方式是什么呢。使用抓包工具抓取请求和响应的报文，用ai计算了token消耗也对不上
What steps can reproduce the bug?
不好复现，我使用抓包工具抓取请求和响应的报文，用ai计算了token消耗
What is the expected behavior?
No response
Additional information
No response
What version of Kimi Code CLI is running?
1.37.0
Which open platform/subscription were you using?
kimi code plan 99
Which model were you using?
kimi 2.6
What platform is your computer?
No response
What issue are you seeing?
The token was consumed too quickly, and errors were reported many times, but the token was also consumed. What is the current calculation method? Use a packet capture tool to capture the request and response packets, and use AI to calculate the token consumption, which is not enough.
What steps can reproduce the bug?
It’s not easy to reproduce. I used a packet capture tool to capture the request and response packets, and used AI to calculate the token consumption.
What is the expected behavior?
No response
Additional information
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No source file, test, or entry point is named. Start by locating the CLI's token-accounting path and comparing its calculations with captured request and response data, including failed requests. Done means the accounting behavior is explained or corrected and the relevant error case is covered by a regression test.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- cli
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100