Simon Mo
Simon Mo
### 🚀 The feature, motivation and pitch Great feedback from one of our user: > For our production monitoring, it'd be great to have more operational metrics for us to...
### Anything you want to discuss about vllm. It would be great if we can distribute nightly images and wheels. This should be as simple as building images to push...
To catch #4304 early.
### Anything you want to discuss about vllm. Currently we run all CI tests matrix on every single commit in pull requests. The CI cost of the vLLM has been...
Fix the docs: https://docs.vllm.ai/en/latest/models/performance.html * Typo * Code
This document includes the features in vLLM's roadmap for Q2 2024. Please feel free to discuss and contribute to the specific features at related RFC/Issues/PRs and add anything else you'd...
The cache was not used previously.
# Tutorial Todos ## Due - [ ] Draft by Friday so Dan can make some edits. - [ ] Test possible collision with Integration (Hari) ## Split into 3...
https://web.mit.edu/decima/ *Learning Scheduling Algorithms for Data Processing Clusters* Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms. Current systems use simple, generalized heuristics and ignore workload characteristics,...