NAME
DeepSeek-Reasonix โ DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability โ leave it running.
SYNOPSIS
INFO
DESCRIPTION
DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability โ leave it running.
README
English ย ยทย ็ฎไฝไธญๆ ย ยทย Website ย ยทย Architecture ย ยทย Benchmarks
A DeepSeek-native AI coding agent for your terminal.
Engineered around prefix-cache stability โ so token costs stay low across long sessions, and you can leave it running.
[!TIP] Cache stability isn't a feature you turn on; it's an invariant the loop is designed around. That's the whole reason Reasonix is DeepSeek-only โ every layer is tuned to the byte-stable prefix-cache mechanic.
[!NOTE] Real user, single day (2026-05-01): 435M input tokens, 99.82% cache hit, ~$12 instead of the ~$61 the same workload would cost with no cache on
v4-flashโ see the case study. DeepSeek provides the cacheable bytes; the four mechanisms in Pillar 1 are how Reasonix keeps them cacheable across long sessions.
Install
cd my-project
npx reasonix code # paste a DeepSeek API key on first run; persists after
Requires Node โฅ 22. Tested on macOS ยท Linux ยท Windows (PowerShell ยท Git Bash ยท Windows Terminal). Get a DeepSeek API key โ ยท reasonix code --help for flags.
What makes Reasonix different
The loop is organized around four pillars. Each one solves a problem generic agent frameworks don't even see โ because they were designed for a different cache mechanic.
Click any card to read the full architecture writeup โ Pillar 1 ยท Pillar 2 ยท Pillar 3 ยท Pillar 4
Capabilities
How it compares
| Reasonix | Claude Code | Cursor | Aider | |
|---|---|---|---|---|
| Backend | DeepSeek | Anthropic | OpenAI / Anthropic | any (OpenRouter) |
| License | MIT | closed | closed | Apache 2 |
| Cost profile | low per task | premium | subscription + use | varies |
| DeepSeek prefix-cache | engineered | not applicable | not applicable | incidental |
| Embedded web dashboard | yes | โ | n/a (IDE) | โ |
| Persistent per-workspace sessions | yes | partial | n/a | โ |
| Plan mode ยท MCP ยท hooks ยท skills | yes | yes | yes | partial |
| Open community development | yes | โ | โ | yes |
For live cache-hit rates, costs, and methodology, see benchmarks/ โ the numbers move with model pricing, so they live with the harness, not in the README.
Documentation
- Architecture โ the four pillars, cache-first loop, harvesting, scaffolds
- Benchmarks โ ฯ-bench-lite harness, transcripts, cost methodology
- Website โ getting started, dashboard mockup, TUI mockup
- Contributing โ comment policy, error-handling rules, library-over-hand-rolled
- Code of Conduct ยท Security policy
Community
[!NOTE] Reasonix is open source and community-developed. The contributors wall below isn't decoration โ every avatar is a real PR that shipped.
Scoped starter tickets โ each with background, code pointers, acceptance criteria, and hints โ live under the good first issue label. Pick anything open.
Open Discussions โ opinions wanted:
- #20 ยท CLI / TUI design โ what's broken, what's missing, what would you change?
- #21 ยท Dashboard design โ react against the proposed mockup
- #22 ยท Future feature wishlist โ what would you build into Reasonix next?
Already using Reasonix and willing to help others discover it? Publish blog posts, articles, screenshots, talks, or videos to Show and tell. The project has no marketing budget โ community word of mouth is how new users find it. Sustained advocates earn the badge below, displayed next to the contributors wall once awarded:
Before your first PR: read CONTRIBUTING.md โ short, strict rules (comments, errors, libraries-over-hand-rolled). tests/comment-policy.test.ts enforces the comment ones; npm run verify is the pre-push gate. By participating you agree to the Code of Conduct. Security issues โ SECURITY.md.
Non-goals
[!IMPORTANT] Reasonix is opinionated. Some things it deliberately doesn't do โ listed here so you can pick the right tool for your work.
- Multi-provider flexibility. DeepSeek-only on purpose. Coupling to one backend is the feature, not a limitation.
- IDE integration. Terminal-first. The diff lives in
git diff, the file tree inls. The dashboard is a companion, not a Cursor replacement. - Hardest-leaderboard reasoning. Claude Opus still wins some benchmarks. DeepSeek is competitive on coding; if your work is "solve this PhD proof" rather than "fix this auth bug," start with Claude.
- Air-gapped / fully-free. Reasonix needs a paid DeepSeek API key. For air-gapped or zero-cost runs see Aider + Ollama or Continue.
Star History
MIT โ see LICENSE
Built by the community at esengine/reasonix