DeepCode vs open-swe
The two are close in size: 16.5k stars for DeepCode, 10.7k for open-swe. Over the days we have tracked them DeepCode moved +4.7% and open-swe +7.5%, so open-swe is growing faster right now.
Neither one leads on the six capability axes, so the choice comes down to which of them fits the way you already work.
Stars and commit dates come from our own daily tracking. The six axes are read off each project's documentation by our review pipeline, so they describe what a project says about itself, not what we measured in its code.
Where they stand today
DeepCode pairs multi-agent delegation and reusable SKILL playbooks with autonomous, test-driven loops and parallel workers so it can decompose large features, run them in isolation, and iterate until tests pass.
- Stars
- 16.5k
- Tracked growth
- +4.7%
- Maturity
- ●●●●●
- Last commit
- 4d ago
- Language
- Python
- License
- MIT
- Cost to run
- Requires your model API key (OpenRouter or other provider)
Provides pluggable isolated cloud sandboxes and subagent orchestration out of the box for safe, parallelized repo edits and automatic PR creation.
- Stars
- 10.7k
- Tracked growth
- +7.5%
- Maturity
- ●●●●●
- Last commit
- 2h ago
- Language
- Python
- License
- MIT
- Cost to run
- Your API key + cloud sandbox costs
Six axes, head to head
Each axis runs 0 to 5. The label under a score is what that project's own docs claim, not a category average.
| Axis | DeepCode | open-swe |
|---|---|---|
Context depth How much of your codebase it sees before it answers: the open diff, the diff plus related files, or the whole repository. | ●●●●● Whole-repo analysis | ●●●●● Whole-repo access |
Noise control How it keeps output volume down — severity thresholds, deduplication, incremental runs over new commits only. | ●●●●● Permission & validation | ●●●●● Basic filters & middleware |
Customization How far it bends to your team: custom rules, prompts, style guides, per-path config. | ●●●●● Extensible skills & hooks | ●●●●● Highly customizable |
Privacy Whether your code stays on your own infrastructure: fully local, self-hostable, or cloud API only. | ●●●●● BYO API key | ●●●●● Cloud APIs (your key) |
Model freedom Whether you can point it at any provider, or it is wired to one. | ●●●●● Provider-agnostic (BYOK) | ●●●●● Bring-your-own-key |
Setup ease What it takes to get a first useful run out of it. | ●●●●● Config + API key | ●●●●● Multi-service setup |
Which one to pick
Pick DeepCode if…
Autonomous test-driven loops — pick DeepCode when you want an agent that plans, delegates across parallel workers, and iterates on a real repository until the project's tests are green.
Pick open-swe if…
Customizable architecture — swap sandboxes, models, tools, and middleware to build an internal coding agent that runs each task in an isolated cloud sandbox and opens PRs automatically.
What people want from each one
HKUDS/DeepCode
Hacker News: DeepCode: Open Agentic Coding drew 18 points and 2 comments.
langchain-ai/open-swe
Hacker News: An Open-Source Asynchronous Coding Agent drew 5 points and 0 comments.
Questions people ask
Is DeepCode better than open-swe?
Neither one leads on the six capability axes, so the choice comes down to which of them fits the way you already work. DeepCode is worth picking when autonomous test-driven loops — pick DeepCode when you want an agent that plans, delegates across parallel workers, and iterates on a real repository until the project's tests are green.
Which of DeepCode and open-swe keeps my code private?
DeepCode: BYO API key (3/5). open-swe: Cloud APIs (your key) (3/5).
What does each one cost to run?
DeepCode: Requires your model API key (OpenRouter or other provider). open-swe: Your API key + cloud sandbox costs.
Full profiles: HKUDS/DeepCode and langchain-ai/open-swe. Everything else in Autonomous SWE agents.