#9292nd of 143 in Coding agents and AI editors
little-coder
Terminal coding agent tuned for small local models, built on pi
- Stars
- 2.7k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Last release
- Sep 2026
- Language
- TypeScript
1 on 7 Apr 20262,697 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
little-coder is a terminal coding agent that adds about 30 extensions and 30 skill markdown files to the pi agent runtime. It targets small local models, with Qwen3.6-35B-A3B on llama.cpp as the default, and also works with Ollama, LM Studio, other OpenAI-compatible servers and cloud providers. It includes plan mode, deep research, read-only sub-coders, background jobs and per-phase model selection.
Who it is for: Developers running coding agents on local models on a laptop or LAN GPU box
Strengths
- Default setup runs Qwen3.6-35B-A3B on 8 GB VRAM by keeping MoE experts in RAM
- Works with llama.cpp, Ollama, LM Studio, any OpenAI-compatible server and cloud providers
- Small-model extensions auto-disable for large or cloud models; cold-start context is about 7k tokens
- Separate plan and action models via /plan-model and /action-model; read-before-edit enforced
Weaknesses
- Requires Node.js 22.19+ even when installed with bun
- Globally installed pi packages do not load unless you opt in via flag or env var
- Dense Qwen3.8-27B measured about 6.4 tok/s on 8 GB VRAM versus about 44 for the MoE
- Local setup means building llama.cpp and downloading a roughly 22 GB model yourself
What it needs
- GPU optional
- Needs Node.js 22.19+, pi, llama.cpp, Ollama or LM Studio (for local models)
- Models: llama.cpp (GGUF), Ollama, LM Studio, OpenAI-compatible servers, Anthropic
Also in Coding agents and AI editors
See all 143| Rank | Project | Score |
|---|---|---|
| 1 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 84 out of 100 |
| 2 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 3 | Open SWEAgent that turns issues into pull requests, reviews PRs and monitors CI | 83 out of 100 |
| #4 | piTerminal coding agent with extensions, skills, and an SDK | 80 out of 100 |
| #5 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |