Skip to content

Conversation

@zhulinJulia24
Copy link
Collaborator

  1. add v1.8 datasets
  2. add api rollout testcase
  3. support num_gpus params in altas datasets
  4. add --dump-res-length params in test and update baselines
  5. use bs = 1 in vllm infer backend because of response is changing when bs > 1

@Myhs-phz Myhs-phz merged commit d836b49 into open-compass:main Dec 15, 2025
8 checks passed
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.

3 participants