microsoft / microsoft/MInference

[Bug]: Implement of TriangleMix

Open
#183 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Python
Stars
1.2k
Forks
82
Avg merge
1d 18h
Merged PRs (30d)
1

Description

Describe the bug

When I try to apply tri_mix on qwen2.5-7B, I meet the error: Failed to apply TriangleMix: The attn_type tri_mix you specified is not supported.

I test with MInferenceConfig.get_available_attn_types(), but also miss the tri_mix.

How can I apply tri_mix on my LLM?

>>> from minference import MInferenceConfig
To benefit from fast kernel implementations, we recommend installing SGLang or vllm.
>>> supported_attn_types = MInferenceConfig.get_available_attn_types()
>>> supported_attn_types
['minference', 'vllm_minference', 'hf', 'vllm', 'dense', 'static', 'dilated1', 'dilated2', 'a_shape', 'tri_shape', 'vllm_a_shape', 'vllm_tri_shape', 'inf_llm', 'flexprefill', 'vllm_flexprefill', 'xattention']
>>> 
Steps to reproduce

No response

Expected Behavior

No response

Logs

No response

Additional Information

No response

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at MInferenceConfig.get_available_attn_types() and trace the code path that rejects attn_type=tri_mix. Confirm the intended TriangleMix integration for qwen2.5-7B; done means tri_mix is accepted, listed as available, and can be applied without the reported error.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.