Skip to content

Pull requests: vllm-project/vllm

New pull request New
599 Open 7,827 Closed
599 Open 7,827 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[ROCm] Effort to reduce the number of environment variables in command line ci/build
#17229 opened Apr 26, 2025 by hongxiayang Loading…
1
Use CUDA 12.6 as default for release and nightly wheels ci/build documentation Improvements or additions to documentation
#17224 opened Apr 26, 2025 by huydhn • Draft
1
[Bugfix] Get a specific type of layer from forward context tpu Related to Google TPUs v1
#17222 opened Apr 26, 2025 by heheda12345 Loading…
1
[Doc] Clarify note for H2O-VL documentation Improvements or additions to documentation
#17219 opened Apr 26, 2025 by DarkLight1337 Loading…
1
[V1][Spec Decode] Apply torch.compile & cudagraph to EAGLE documentation Improvements or additions to documentation v1
#17211 opened Apr 26, 2025 by luyuzhe111 Loading…
2
[Misc][Tools][Benchmark] Publish script to auto tune server parameters
#17207 opened Apr 25, 2025 by Chenyaaang Loading…
2
[Benchmark] Add single turn MTBench to Serving Bench
#17202 opened Apr 25, 2025 by ekagra-ranjan Loading…
2
[Misc]add configurable cuda graph size
#17201 opened Apr 25, 2025 by CXIAAAAA Loading…
3
[Hardware][Apple] Allows VLLM_TARGET_DEVICE=empty on MacOs ci/build ready ONLY add when PR is ready to merge/full CI is needed
#17200 opened Apr 25, 2025 by wallashss Loading…
1
[WIP] Support vLLM in transformers hybrid attention implementation
#17198 opened Apr 25, 2025 by wuisawesome Loading…
1
[Security] Don't bind tcp zmq socket to all interfaces documentation Improvements or additions to documentation security Security related issues and PRs
#17197 opened Apr 25, 2025 by russellb Loading… v0.8.5
2
[Bugfix] Fix Lora Name Parsing
#17196 opened Apr 25, 2025 by alex-jw-brooks Loading…
2
[WIP][Bugfix] Fix 'MistralTokenizer' object has no attribute 'init_kwargs' bug Something isn't working ready ONLY add when PR is ready to merge/full CI is needed
#17195 opened Apr 25, 2025 by chaunceyjiang Loading… v0.8.5
1
7
[V1] Remove num_input_tokens from attn_metadata tpu Related to Google TPUs v1
#17193 opened Apr 25, 2025 by heheda12345 Loading…
1
[Bugfix] support local dataset path in benchmark_serving
#17179 opened Apr 25, 2025 by wubai Loading…
1
Add option "--expand-tools-even-if-tool-choice-none" frontend tool-calling
#17177 opened Apr 25, 2025 by okdshin Loading…
1
[CI] Add mteb testing to test the accuracy of the embedding model ci/build
#17175 opened Apr 25, 2025 by noooop Loading…
5
[Bugfix] Modifications to error handling of multiple vllm api endpoints frontend
#17165 opened Apr 25, 2025 by tunglinwood Loading…
7
[Hardware][Power] Enable compressed tensor W8A8 INT8 quantization for POWER ci/build
#17153 opened Apr 25, 2025 by Akashcodes732 Loading…
1
[Misc] Add gemma3 chat template with pythonic-style function calling documentation Improvements or additions to documentation tool-calling
#17149 opened Apr 25, 2025 by philipchung Loading…
1
4
Add xLAM tool parser support documentation Improvements or additions to documentation frontend tool-calling
#17148 opened Apr 25, 2025 by zuxin666 Loading…
6
[Model] Mamba2 causal conv1d Refactor to Split Prefill and Decode Requests for Corresponding Kernels
#17146 opened Apr 25, 2025 by cyang49 • Draft
2
[Frontend][TPU] Enforce user input key args to reduce chance of large performance degradation documentation Improvements or additions to documentation frontend
#17145 opened Apr 24, 2025 by Chenyaaang Loading…
4
[Kernel] FP8 quantization fused into V1 Triton Attention
#17143 opened Apr 24, 2025 by gshtras Loading…
1
[ROCm][FP8][Kernel] FP8 quantization fused into Custom Paged Attention
#17139 opened Apr 24, 2025 by gshtras Loading…
6
ProTip! Follow long discussions with comments:>50.