killi — a fast, tiny AI coding agent for your terminal
Written in Rust · one ~7 MB native binary · ~18 MB RAM · 70–90% prompt-cache hits
Works with ChatGPT, Claude and any OpenAI/Anthropic-compatible LLM · MCP, subagents, plan mode, skills, hooks, AGENTS.md
Apple silicon Macs · MIT licensed · GitHub
Benchmarks
Measured with scripts/bench.py on an M4 Pro (killi 0.6.56, Claude Code 2.1.319, 2026-11-07; Kimi Code 2.3.0 and opencode 2.1.3, October 2026). Details in docs/bench.json.
| killi | Claude Code | Kimi Code | opencode 2 | |
|---|---|---|---|---|
| Installed size | ~7 MB | 307 MB | 188 MB | 204 MB |
| Startup | < 10 ms | 72 ms | 369 ms | 77 ms |
| Memory, one prompt | ~18 MB | 318 MB | 408 MB | 332 MB |
| System prompt + tools | ~2,600 tokens | — | — | — |
| Input tokens, same 14 evals | 1× | — | 5.6× | — |
| Prompt-cache hits | 70–90% | — | — | — |
killi passed every capability task Kimi Code passed on the same model, with 5.6× fewer input tokens. See docs/eval.md.
Features
- Small and quick. One 7 MB binary, no runtime; starts in under 10 ms.
- Few tokens. ~2,600-token system prompt and tools; 5.6× fewer input tokens than Kimi Code on the same evals.
- Cache-friendly by design. Fixed system prompt and tool list, append-only history: 70–90% of input tokens come from the provider's prompt cache.
- Any LLM. OpenAI-compatible, Anthropic-compatible or OpenAI Responses; ChatGPT, OpenCode Go, Muse Code and Command Code subscriptions.
- Your setup, carried over.
killi import, skills, MCP servers, Claude Code-style hooks and AGENTS.md. - MCP, subagents, plan mode, undo. Up to 128 parallel subagents, plan mode, background tasks, /undo.
FAQ
What is killi?
killi is a fast, tiny, open-source AI coding agent for the terminal, written in Rust. It ships as one ~7 MB native binary for Apple silicon Macs with no runtime to install, starts in under 10 ms, and uses about 18 MB of memory while running a prompt. Like Claude Code, Codex CLI, opencode or Kimi Code, it reads your code, runs commands and edits files, but it is free (MIT), works with any OpenAI-compatible, Anthropic-compatible or OpenAI Responses LLM, and is built so the provider's prompt cache keeps hitting.
Is killi open source?
Yes. killi is open source under the MIT license, and the full source is on GitHub at github.com/timwuhaotian/killi. It is free to use and installs as a single ~7 MB binary.
How is killi different from Claude Code?
Claude Code is Anthropic's closed-source agent tied to Anthropic models. killi is open source, model-agnostic, and much lighter: in the same benchmark, killi used about 18 MB of memory while Claude Code used 318 MB, Kimi Code 408 MB and opencode 2 332 MB. killi passed every capability task Kimi Code passed on the same model with 5.6× fewer input tokens, and 70–90% of its input tokens are served from the provider's prompt cache.
How much memory does killi use?
About 18 MB of peak memory while running a prompt, measured on 2026-11-07 on an M4 Pro with scripts/bench.py (hooks disabled). For comparison, the same benchmark measured Kimi Code at 408 MB, Claude Code at 318 MB and opencode 2 at 332 MB. The installed binary itself is about 7 MB.
Does killi support ChatGPT?
Yes. killi login chatgpt signs in with a ChatGPT Plus or Pro account in the browser, through OpenAI's official sign-in for open-source tools. Requests go to api.openai.com/v1/responses on your plan, count toward your ChatGPT usage, and can be monitored in ChatGPT settings → Usage. killi also works with OpenAI-compatible APIs, Claude (Anthropic), and many other providers.
Does killi support Claude / Anthropic models?
Yes. killi accepts an Anthropic API key and speaks the Anthropic Messages protocol, so Claude models work directly. Claude models behind an OpenAI-compatible gateway can be used with the Responses protocol, with cache_control support for prompt caching.
What is prompt caching and why does killi care?
Providers cache the start of a prompt, so a request that begins exactly like the previous one is cheaper and faster. killi is built around this: the system prompt, the tool list and the history are fixed for the whole session and only appended to, so 70–90% of input tokens come from the cache, including in plan mode, subagents, swarms and resumed sessions.
Does killi support MCP servers, subagents and plan mode?
Yes. killi supports MCP servers (stdio and HTTP) from mcp.json, subagents and swarms (up to 128 in parallel), plan mode that blocks edits until you approve, Claude Code-style hooks, AGENTS.md / CLAUDE.md instructions, skills, undo, background tasks and web search.