Set a local quantized GPU ONXX model to run on RTX GPU
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 812
- Forks
- 175
- Avg merge
- 2m
- Merged PRs (30d)
- 3
Description
Sorry this is a bit new for me, but I'd like to incorporate local models like https://huggingface.co/nvidia/Mistral-7B-Instruct-v0.3-ONNX-INT4 or similar models to run locally instead. I think the token speed could be dramatically improved with this quantized models on the GPU rather than using a CPU version.
I can take a look in Cursor / Gemini Code, if that would help, but want to run it by you first, @ruvnet .
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
No files, tests, or entry points are named. Start by locating the current local-model and CPU inference path, then determine whether an RTX-compatible ONNX runtime path is supported. Done would require a defined local quantized-model integration and evidence that inference runs on the GPU with improved token speed.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- ai, machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100