meta-pytorch / meta-pytorch/KernelAgent
Support DeepSeek as a LLM Provider
Open
@Jack-Khuu is already working on this.
Since Dec 27, 2025.
- Dominant language
- Python
- Stars
- 556
- Forks
- 92
- PR merge metrics
- No merged PRs in 30d
Description
Hi maintainers👋,
There was an earlier implementation in PR #6, which DeepSeek was added but later removed. Can we re-support it as a LLM provider? I know GPT and Claude are strong enough in code generation, but they are restricted in some region. Besides, Deepseek is much cheaper compared to them. Lastly, Deepseek uses an OpenAI-compatible API, which means we don't need to refractor the code.
Happy to hear your feedbacks.
cc @sijiac @Laurawly
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.