Bug report - Invoking deepseek-reasoner via /model deepseek-reasoner causes error

[adunato]

Aider version: 0.72.3
Python version: 3.11.10
Platform: Windows-10-10.0.26100-SP0
Python implementation: CPython
Virtual environment: No
OS: Windows 10 (64bit)
Git version: git version 2.46.0.windows.1

Aider v0.72.3
Model: deepseek-reasoner with diff edit format, infinite output
Git repo: .git with 269 files
Repo-map: using 4096 tokens, auto refresh

litellm.BadRequestError: LLM Provider NOT provided. Pass in the LLM provider you are trying to call. You passed
model=deepseek-reasoner
Pass model as E.g. For 'Huggingface' inference endpoints pass in completion(model='huggingface/starcoder',..) Learn more:
https://docs.litellm.ai/docs/providers

https://docs.litellm.ai/docs/providers

[adunato]

/model deepseek/deepseek-reasoner has worked before, but will often lead to no response.

I think these may be remote issues with the deepseek API but I wonder whether there needs to be more management on the error side / timeout in aider.

[oregapam]

I have same issue, no output when I try to use --model r1 argument and try to prompt something. However the web based chat (deepseek.com) have issues too.

[wladimiiir]

Deepseek API has been experiencing issues for past few days due to some malicious attacks.
You can check the status here: https://status.deepseek.com/