Azure-Samples / Azure-Samples/swiss-llm-quickstart
Apertus 1.5 update
- Dominant language
- Python
- Stars
- 16
- Forks
- 2
- PR merge metrics
- No merged PRs in 30d
Description
The updated model weights of Apertus with new features and better performance are available. We are still finalizing work with the community on driver support. Just want to give you a heads up, hopefully we can update the launch scripts and improve their stability. Please let me know if there's anything you need a hand with. Details for vLLM that I think you'll find particularly relevant can be found on the [model cards](https://huggingface.co/swiss-ai/Apertus-v1.5-70B) for 8B / 70B and our [documentation area](https://apertus-ai.org/docs/guides/vllm/).
Thanks for your work on this so far @faxg @dbroeglin
Contributor guide
Research direction
Start by reviewing the repository's launch scripts, then read the linked Apertus 1.5 model cards and vLLM documentation. Determine which scripts need updating for the 8B and 70B models and how stability should be improved; driver support is still being finalized with the community, so confirm its scope before starting.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- ai, devops
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100