[Vulkan] Improve latency for non-LLM models
Open
Nobody has claimed this yet.
module: vulkan
- Dominant language
- Python
- Stars
- 5k
- Forks
- 1.2k
- Avg merge
- 2d 10h
- Merged PRs (30d)
- 581
Description
On par latency for MV3, ResNet50 on S24, Pixel8-pro, Pixe7-pro to ET + XNNPACK
cc @SS-JIA @manuelcandales @cbilgin
Contributor guide
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 identifying the Vulkan path used for MV3 and ResNet50, then measure latency on the S24, Pixel 8 Pro, and Pixel 7 Pro. Compare the results with ET + XNNPACK; done means non-LLM model latency is on par across the listed devices.
Written by the indexing model from the issue text.
Assessment
- Domain
- machine-learning, mobile-dev, performance
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100