vllm-metal

Python Medium

vllm-project/vllm-metal

738 stars

76 forks

8 open issues

Active Mar 2026

Beginner-Friendly Issues 8

Issues tagged for new contributors

View all on GitHub

Server stalls mid-stream for large context requests, never sends [DONE] or closes connection

#199 · Mar 22, 2026

Medium

good first issue

[Qwen3.5] Tracking issue & roadmap on Qwen3.5 collaboration

#194 · Mar 22, 2026

Medium

Paged Attention Model Compact Track

#191 · Mar 21, 2026

Medium

[RFC] [kernel] Paged attention kernels should be MLX primitives to eliminate per-layer sync barriers

#188 · Mar 21, 2026

Medium

[RoadMap] [Paged KV] Prefix Caching on Paged KV Cache Path

#182 · Mar 20, 2026

Medium

Check if kernel partition is working or not

#180 · Mar 19, 2026

Medium

When running "vllm bench serve --num-prompts=3" `assert request.num_output_placeholders >= 0`

#171 · Mar 17, 2026

Medium

[RoadMap] [Paged KV] Continuous Batching + Chunked Prefilling + paged varlen flash att

#148 · Mar 9, 2026

Medium

Sign in required

Authenticate to use favourites & bookmarks

5