Messages 为什么会占用这么多 token 啊? 我就发了个 hello which model do you use?
╭─── Claude Code v2.1.114 ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ │ Tips for getting started │
│ Welcome back! │ Run /init to create a CLAUDE.md file with instructions for Claude │
│ │ Note: You have launched claude in your home directory. For the best experience, launch it in a project directory instead. │
│ ▐▛███▜▌ │ ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────── │
│ ▝▜█████▛▘ │ Recent activity │
│ ▘▘ ▝▝ │ No recent activity │
│ │ │
│ MiniMax-M2.7 with medium effo… · API Usage Billing │ │
│ /Users/ │ │
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
❯ hello which model do you use?
∴ Thinking…
The user is asking which model I use. Based on the system prompt, I can see I'm powered by the model MiniMax-M2-M2. I should answer this directly and concisely.
⏺ I'm powered by MiniMax-M2. This conversation is running on that model.
❯ /context
⎿ Context Usage
⛁ ⛁ ⛁ ⛁ ⛁ ⛀ ⛀ ⛁ ⛁ ⛶ MiniMax-M2.7
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ 15.8k/200k tokens (8%)
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ Estimated usage by category
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛁ System prompt: 5.8k tokens (2.9%)
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛁ System tools: 5.3k tokens (2.6%)
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛁ Memory files: 548 tokens (0.3%)
⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛁ Skills: 2.7k tokens (1.4%)
⛶ ⛶ ⛶ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛁ Messages: 2k tokens (1.0%)
⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛶ Free space: 150.7k (75.4%)
⛝ Autocompact buffer: 33k tokens (16.5%)
Memory files · /memory
└ .claude/CLAUDE.md: 548 tokens
Skills · /skills
User
├ request-analyzer: 118 tokens
├ turborepo: 118 tokens
├ git-commit: 108 tokens
├ frontend-design: 104 tokens
├ vercel-react-best-practices: 89 tokens
├ vercel-composition-patterns: 85 tokens
├ vue-best-practices: 83 tokens
├ vercel-react-native-skills: 81 tokens
├ find-skills: 79 tokens
├ planning-with-files: 71 tokens
├ ralph: 64 tokens
├ fmp-job-scheduler-api: 64 tokens
├ skill-creator: 60 tokens
├ karpathy-guidelines: 60 tokens
├ yunxiao-devops: 60 tokens
├ slidev: 58 tokens
├ excalidraw: 55 tokens
├ brainstorming: 53 tokens
├ rlm: 52 tokens
├ pnpm: 48 tokens
├ vitest: 46 tokens
├ vue: 46 tokens
├ pinia: 46 tokens
├ unocss: 46 tokens
├ tsdown: 45 tokens
├ vite: 44 tokens
├ nuxt: 43 tokens
├ vitepress: 42 tokens
├ defuddle: 39 tokens
├ mermaid-diagram-specialist: 34 tokens
├ fmp-commit: 33 tokens
├ vueuse-functions: 29 tokens
└ antfu: 21 tokens
Plugin
└ init-rules: 69 tokens
※ recap: You asked about which model I'm using, I answered MiniMax-M2. No task was started — you were just checking in. (disable recaps in /config)
1 个帖子 - 1 位参与者
来源: linux.do查看原文