Skip to content

refactor: context usage now relay on llm response#64

Open
zhangfeiran wants to merge 15 commits intomindspore-lab:mainfrom
zhangfeiran:v0.1-ctx
Open

refactor: context usage now relay on llm response#64
zhangfeiran wants to merge 15 commits intomindspore-lab:mainfrom
zhangfeiran:v0.1-ctx

Conversation

@zhangfeiran
Copy link
Copy Markdown
Collaborator

@zhangfeiran zhangfeiran commented Apr 9, 2026

  • 默认使用llm response里的total_tokens,而非本地简易tokenizer,仅当模型无返回时才fallback
  • 添加/ctx命令,显示具体模型返回context消耗详情
  • 其他:debug模式下UI显示ctx具体值;mscli --help信息优化;reserved context默认改为10%;python shell输出默认无buffer

已包含 #63 改动
resolves #60 part 2

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.

context related issue

1 participant