claw-code vs codewhale
Two of the top projects in Coding agents and AI editors, side by side: who leads on each score part and who each one suits, then setup, license, activity and what each review found.
claw-code
Rust CLI coding agent harness using Anthropic and OpenAI-compatible APIs
codewhale
Terminal coding agent in Rust that works with 40+ model providers
Verdict
claw-code ranks higher, 97 to 94. Of the 4 score parts we measure for both, claw-code leads on 2, codewhale on none and 2 are even.
- Adoptionclaw-code leads by 8 points, 99 to 91.
- FreshnessEven at 100.
- Maintenanceclaw-code leads by 1 point, 99 to 98: close to even.
- Agent-readyEven at 70.
Pick claw-code if you are
Developers who want to inspect or build a Rust CLI coding agent
- `claw doctor` checks API key, model access and tool configuration after build
- Docs cover local OpenAI-compatible providers: Ollama, llama.cpp, vLLM
Pick codewhale if you are
Developers who want a model-agnostic terminal coding agent with audit trails
- /receipts lists changed files, commands run, web and MCP calls, and approvals
- Plan, Work and Operate modes with Ask, Auto-Review and Full Access permissions
| What we compare | claw-code | codewhale |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 99, widely used | 91, widely used |
| Freshness | 100, active | 100, active |
| Maintenance | 99, healthy | 98, healthy |
| Agent-ready | 70, partly | 70, partly |
| Facts from GitHub and the README | ||
| Stars | 195k | 41.2k |
| License | MIT (permissive) | MIT (permissive) |
| Last commit | Oct 2026 | Nov 2026 |
| Last release | Oct 2026 | Oct 2026 |
| Language | Rust | Rust |
| Docker | Yes | Yes |
| GPU | Not needed | Optional |
| arm64 or Apple Silicon | Mentioned | Mentioned |
claw-code
Claw Code is a Rust implementation of the `claw` command-line coding agent, built from source with cargo. It runs one-shot prompts or interactive sessions using an API key, and the docs cover local OpenAI-compatible servers such as Ollama, llama.cpp and vLLM. The README describes the repo as an agent-managed exhibit, not the production project, and points to LazyCodex and Gajae-Code for real work.
Who it is for: Developers who want to inspect or build a Rust CLI coding agent
Strengths
- `claw doctor` checks API key, model access and tool configuration after build
- Docs cover local OpenAI-compatible providers: Ollama, llama.cpp, vLLM
- Build steps documented for macOS, Linux and Windows PowerShell
- MIT license; workspace test suite and deterministic mock-service parity harness
Weaknesses
- README says it is not the serious production project and points elsewhere
- Build-from-source only; `cargo install claw-code` installs a deprecated stub
- Requires an API key; Claude subscription login is not supported
- No ACP/Zed daemon yet; `claw acp serve` only returns status
- no GPU
- Docker + Compose
- Needs Rust toolchain (cargo), LLM API key
- Models: Anthropic API, OpenAI-compatible APIs, Ollama, llama.cpp, vLLM
codewhale
Codewhale is a terminal coding agent that reads a repository, edits files and runs shell commands until a task is done or it needs input. It supports more than 40 built-in providers, any OpenAI-compatible endpoint, and local models through Ollama, vLLM or SGLang, with DeepSeek as the default. It also has a headless exec mode, a local web UI, a PR review command and an ACP server for editors.
Who it is for: Developers who want a model-agnostic terminal coding agent with audit trails
Strengths
- /receipts lists changed files, commands run, web and MCP calls, and approvals
- Plan, Work and Operate modes with Ask, Auto-Review and Full Access permissions
- One engine behind the TUI, exec, local web UI, Runtime API and ACP over stdio
- Reads AGENTS.md; opt-in import of CLAUDE.md and Cursor rules; supports MCP, skills, hooks
Weaknesses
- Telemetry is on by default; aggregate counts are sent until you disable it
- No OS sandbox on Windows, and Full Access disables the sandbox everywhere
- Some features, such as the one-step /undo and terminal workbar, are only in unreleased 0.10.2
- exec --auto defaults to Full Access with no shell sandbox
- GPU optional
- Docker
- Needs Git repository, Provider API key or local runtime (Ollama, vLLM, SGLang), bubblewrap (optional, Linux sandbox)
- Models: DeepSeek, Anthropic, OpenAI, Google, Mistral
More in Coding agents and AI editors
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.