Issue 2026-06-28
v0.8.x
CodeWhale
any model, open models first
实 时LIVE
PR#3712fix(telegram): render markdown messages safely· 17mPR#3711test(tui): cover unknown tool lifecycle errors· 35mPR#3710Feat/plugin p3 mcp· 3hPR#3708feat(plugins): add manifest parsing, discovery, and registry· 4hPR#3709Feat/plugin p2 cli· 4hISS#3192Put it up for agentclientprotocol/registry· 6hISS#3388v0.8.66 EPIC: Token, cache, and context discipline release gate· 6hPR#3707docs: add v0.8.66 release ledger· 6hISS#3541Feature Request: Rust-based native runtime / desktop client for lower latency · 6hISS#1732合并分析报告保存文档巨慢· 6hISS#1818token消耗超级大· 6hISS#1747Cache hit problem· 6hPR#3712fix(telegram): render markdown messages safely· 17mPR#3711test(tui): cover unknown tool lifecycle errors· 35mPR#3710Feat/plugin p3 mcp· 3hPR#3708feat(plugins): add manifest parsing, discovery, and registry· 4hPR#3709Feat/plugin p2 cli· 4hISS#3192Put it up for agentclientprotocol/registry· 6hISS#3388v0.8.66 EPIC: Token, cache, and context discipline release gate· 6hPR#3707docs: add v0.8.66 release ledger· 6hISS#3541Feature Request: Rust-based native runtime / desktop client for lower latency · 6hISS#1732合并分析报告保存文档巨慢· 6hISS#1818token消耗超级大· 6hISS#1747Cache hit problem· 6h
Open source · multi-model · runs locally

The best available models, for as many people as possible.

CodeWhale is an open source terminal coding agent — a TUI and a CLI. Point it at a model and a project and it gets to work: reading code, making edits, running commands, checking results, planning multi-step tasks, and correcting itself when something fails. It's genuinely agentic — the system prompt is built for it, and long-running tasks with a real plan are the norm. The goal is simple: keep the local terminal workflow current with practical coding-agent research and capabilities.

Where things stand
Recommended right now: GLM 5.2 on high reasoning — working very well in CodeWhale.
OpenAI Codex login is working.
Kimi OAuth is temporarily broken; on the list.
Don't see a model or endpoint you want? Open an issue.
Maintained by Hmbown0.8.6527 providers
one-line install
# npm — no Rust toolchain required
$ npm install -g codewhale
$ codewhale

# other ways (Cargo / Homebrew / direct download / China mirrors): # /install
needs Node or Rust 1.88+other ways →
Stars · 星标
39.1k
Forks · 复刻
3.4k
Contributors · 贡献者
168
Latest · 版本
v0.8.65

Get started

Three steps: install it, point it at a model key, and run it in a project. The install guide walks through each one.

What's going on

Quiet release week — install paths and contributor guides up to date.

This dispatch is regenerated by DeepSeek V4-Flash on a six-hour cron. Until the cron fills in fresh repo activity, the static links below are the best starting points.

  • shipped
    Install

    npm, Cargo, Homebrew, or a direct download.

  • opened
    Open issues

    Start with anything labelled good first issue.

  • discussion
    Roadmap

    What is confirmed, what is being weighed, what is ruled out.

Need help?

v0.8.65 credits

Every patch and report counts

The project

Developers from all over have made CodeWhale what it is.

CodeWhale started as one person's DeepSeek side project, and contributors from many countries have shaped it into what it is — the contributor list on every release is the proof. No CLA, no sponsor lockouts, issues triaged in the open, releases cut from main.

“All feedback is a gift.” Something I learned early in teaching, and the way I maintain CodeWhale. Every issue, PR, bug report, or idea is how I understand real use cases and make better decisions.

If you hit something that doesn't work, or want a model that isn't listed — that's the most useful thing you can tell the project.

version0.8.65
providers27
licenseMIT