Skip to content

Pull requests: ROCm/vllm

New pull request New
20 Open 531 Closed
20 Open 531 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Disable skynny gemms by default
#568 opened Jun 5, 2025 by k-artem Loading…
1
Patch to run AITER 0507
#541 opened May 8, 2025 by qli88 Loading…
Remap fp8 kv-scale names for Deepseek
#535 opened May 1, 2025 by sstamenk Loading…
1
Updated README.md with April 29 results
#526 opened Apr 27, 2025 by Mcirino1 Loading…
7
BF16 Skinny Optimization
#520 opened Apr 22, 2025 by amd-hhashemi Loading…
Enable RPD Profiler in OpenAI server
#513 opened Apr 15, 2025 by rebklee Loading…
1
removing quant and kv-cache fp8 from deepseek run instructions
#509 opened Apr 9, 2025 by arakowsk-amd Loading…
5
creating 1 gpu agents on OCI cluster
#488 opened Mar 24, 2025 by dhonnappa-amd Loading…
Handling input dim size greater than 3 in tuned_gemm.py
#482 opened Mar 13, 2025 by charlifu Loading…
EXPERIMENTING WITH K8S // NO NEED TO MERGE // Rocm vllm ci fix nd k8 osci
#477 opened Mar 12, 2025 by Alexei-V-Ivanov-AMD Loading…
Rocm vllm ci fix
#468 opened Mar 10, 2025 by Alexei-V-Ivanov-AMD Loading…
Test Queues
#456 opened Feb 28, 2025 by dhonnappa-amd • Draft
Enable custom paged attention kernel for Navi 3/4
#446 opened Feb 24, 2025 by hyoon1 Loading…
1
updating dev-docker README 20250214
#426 opened Feb 14, 2025 by arakowsk-amd • Draft
Updating ISL and OSL to align with reported benchmark table stale
#424 opened Feb 14, 2025 by eduand-alvarez Loading…
1
K8test baseline -> Testing a single MI300 8x GPU node for CI performance // no need to merge stale
#409 opened Feb 6, 2025 by Alexei-V-Ivanov-AMD Loading…
1
K8 node testing // no need to merge stale
#404 opened Feb 4, 2025 by Alexei-V-Ivanov-AMD Loading…
1
Fp8 header stale
#396 opened Jan 31, 2025 by gshtras Loading…
1
[Bugfix] Deepseek v3 fix max_num_batched_tokens
#386 opened Jan 24, 2025 by Concurrensee • Draft
merge paged attention feature and moe feature into llama_fp8_12062024
#370 opened Jan 21, 2025 by yuzho-amd • Draft
ProTip! Adding no:label will show everything without a label.