OpenBMB / OpenBMB/MiniCPM-o-Demo
omni full duplex 卡顿 & torch compile
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 386
- Forks
- 81
- Avg merge
- 1h 59m
- Merged PRs (30d)
- 3
Description
为什么这个例子说话很流程
https://35.226.63.1:8008/omni
是使用了 flash_attention_2 方式么
目前我是使用sdpa 方式，没有https://35.226.63.1:8008/omni流畅
Contributor guide
No contributing guide indexed for this repository
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.
Research direction
Start by reproducing the full-duplex behavior at https://35.226.63.1:8008/omni and compare it with the reported SDPA setup. Investigate whether the demo uses flash_attention_2 or torch.compile, then document or isolate the configuration responsible for the smoothness difference; no source files or tests are named in the issue.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 32/100