Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adapt for DeepSeek-R1 #54

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Conversation

StephenXie
Copy link

@StephenXie StephenXie commented Jan 30, 2025

  • Added support for deepseek-reasoner
  • Requires DEEPSEEK_API_KEY in .env
  • Successfully ran inference_and_check (example of running inference with deepseek-reasoner on APPS dataset)
    • python skythought/tools/inference_and_check.py --dataset APPS --model deepseek-reasoner --tp 2 --max_tokens 8192 --split test --source all --end 50 --result-dir ./data --inference
  • Ran eval.py on AIME and GPDADiamond and accuracy match official results
    • python eval.py --model deepseek-reasoner --evals=AIME,GPQADiamond --tp=2 --output_file=results.txt --temperatures 0.7
      image
      image

@kouroshHakha
Copy link
Collaborator

Thanks @StephenXie for your contribution, but the foundation to be able to add other non-openai api clients is not setup for this repo yet. We need to first finish a few refactoring efforts (like #47) that we have in line before we can add more models through api.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants