[Core] feat: Implement Priority Scheduling in V1 Engine#18700

Closed

amitm02 wants to merge 256 commits intovllm-project:mainfrom

amitm02:feat/v1-priority-scheduling

This pull request is big! We're only showing the most recent 250 commits

Commits on Jun 1, 2025

[Misc] small improve (#18680)

reidliu41
authored and
amitm02
committed
[Bugfix] Fix profiling dummy data for Pixtral (#18677)

DarkLight1337
authored and
amitm02
committed
feat: Implement Priority Scheduling in V1 Engine

google-labs-jules[bot]
authored and
amitm02
committed
Update test_scheduler_priority.py
amitm02
committed
Update request.py
amitm02
committed
pre-commit styling
amitm02
committed
fix types issues as self.waiting can be both deque or list
amitm02
committed
fix type issues of skipped_waiting_requests
amitm02
committed
Update scheduler.py
amitm02
committed
lint fixes
amitm02
committed
fix types issues as self.waiting can be both deque or list
amitm02
committed
fix type issues of skipped_waiting_requests
amitm02
committed
Update scheduler.py
amitm02
committed
lint fixes
amitm02
committed
fix tests
amitm02
committed
Update test_scheduler.py
amitm02
committed
[Core][Multimodal] Convert PIL Image to array without data copy when hashing (#18682)

lgeiger
authored and
amitm02
committed
[CI/Build][Doc] Update `gte-Qwen2-1.5B-instruct` usage (#18683)

authored and
amitm02
committed
[Misc] Fixed the abnormally high TTFT issue in the PD disaggregation example (#18644)

authored and
amitm02
committed
refactor: simplify request handler, use positive condition check for handler assignment (#18690)

googs1025
authored and
amitm02
committed
[Bugfix] Fix the lm_head in gpt_bigcode in lora mode (#6357)

maxdebayser
authored and
amitm02
committed
[CI] add missing argument (#18694)

andyxning
authored and
amitm02
committed
[GH] Add issue template for reporting CI failures (#18696)

DarkLight1337
authored and
amitm02
committed
[Doc] Fix issue template format (#18699)

DarkLight1337
authored and
amitm02
committed
[Bugfix] Fix Mistral-format models with sliding window (#18693)

DarkLight1337
authored and
amitm02
committed
[CI/Build] Replace `math.isclose` with `pytest.approx` (#18703)

DarkLight1337
authored and
amitm02
committed
[CI] fix dump_input for str type (#18697)

andyxning
authored and
amitm02
committed
[Model] Add support for YARN in NemotronNAS models (#18427)

Naveassaf
authored and
amitm02
committed
[CI/Build] Split pooling and generation extended language models tests in CI (#18705)

Isotr0py
authored and
amitm02
committed
[Hardware][Intel-Gaudi] [CI/Build] Add tensor parallel size = 2 test to HPU CI (#18709)

ldurejko
authored and
amitm02
committed
[Misc] add AutoGen integration (#18712)

authored and
amitm02
committed
[Bugfix]: handle hf-xet CAS error when loading Qwen3 weights in vLLM (#18701)

YanWuHao
authored and
amitm02
committed
[Doc] Improve API docs (#18713)

DarkLight1337
authored and
amitm02
committed
[Doc] Move examples and further reorganize user guide (#18666)

DarkLight1337
authored and
amitm02
committed
[Bugfix] Fix Llama GGUF initialization (#18717)

DarkLight1337
authored and
amitm02
committed
[V1][Sampler] Improve performance of FlashInfer sampling by sampling logits instead of probs (#18608)

lgeiger
authored and
amitm02
committed
Convert `examples` to `ruff-format` (#18400)

hmellor
authored and
amitm02
committed
[Model][Gemma3] Simplify image input validation (#18710)

lgeiger
authored and
amitm02
committed
[Misc] improve web section group title display (#18684)

reidliu41
authored and
amitm02
committed
[V1][Quantization] Add CUDA graph compatible v1 GGUF support (#18646)

Isotr0py
authored and
amitm02
committed
[Model][Gemma3] Cast image pixel values already on CPU (#18732)

lgeiger
authored and
amitm02
committed
[FEAT] [ROCm] Upgrade AITER Fused MoE kernels. (#18271)

vllmellm
authored and
amitm02
committed
[Doc] Update OOT model docs (#18742)

DarkLight1337
authored and
amitm02
committed
[Doc] Update reproducibility doc and example (#18741)

DarkLight1337
authored and
amitm02
committed
[Misc] improve docs (#18734)

reidliu41
authored and
amitm02
committed
feat(rocm-support): support mamba2 on rocm (#18565)

almersawi
authored and
amitm02
committed
[Hardware][Intel-Gaudi] [CI/Build] Fix multiple containers using the same name in run-hpu-test.sh (#18752)

ldurejko
authored and
amitm02
committed
[Doc] cleanup deprecated flag for doc (#18715)

calvin0327
authored and
amitm02
committed
Minor fix about MooncakeStoreConnector (#18721)

maobaolong
authored and
amitm02
committed
[Build] fix cpu build missing libtbbmalloc.so (#18744)

kebe7jun
authored and
amitm02
committed
[BUG FIX] minicpm (#18739)

authored and
amitm02
committed
[Doc] Convert Sphinx directives ( `{class}`, `{meth}`, `{attr}`, ...) to MkDocs format for better documentation linking (#18663)

Zerohertz
authored and
amitm02
committed
[CI/Build] Remove imports of built-in `re` (#18750)

DarkLight1337
authored and
amitm02
committed
[V1][Metrics] Add API for accessing in-memory Prometheus metrics (#17010)

markmc
authored and
amitm02
committed
[Misc] Fixed the abnormally high TTFT issue in the PD disaggregation example (#18644)

authored and
amitm02
committed
[GH] Add issue template for reporting CI failures (#18696)

DarkLight1337
authored and
amitm02
committed
[Doc] Fix issue template format (#18699)

DarkLight1337
authored and
amitm02
committed
[Model] Add support for YARN in NemotronNAS models (#18427)

Naveassaf
authored and
amitm02
committed
[Hardware][Intel-Gaudi] [CI/Build] Add tensor parallel size = 2 test to HPU CI (#18709)

ldurejko
authored and
amitm02
committed
[Doc] Improve API docs (#18713)

DarkLight1337
authored and
amitm02
committed
[Doc] Move examples and further reorganize user guide (#18666)

DarkLight1337
authored and
amitm02
committed
[Bugfix] Fix Llama GGUF initialization (#18717)

DarkLight1337
authored and
amitm02
committed
Convert `examples` to `ruff-format` (#18400)

hmellor
authored and
amitm02
committed
[Misc] improve docs (#18734)

reidliu41
authored and
amitm02
committed
[Hardware][Intel-Gaudi] [CI/Build] Fix multiple containers using the same name in run-hpu-test.sh (#18752)

ldurejko
authored and
amitm02
committed
[V1][Metrics] Add API for accessing in-memory Prometheus metrics (#17010)

markmc
authored and
amitm02
committed
Disable prefix cache by default for benchmark (#18639)

cascade812
authored and
amitm02
committed
optimize get_kv_cache_torch_dtype (#18531)

chunxiaozheng
authored and
amitm02
committed
[Core] Automatically cast multi-modal input dtype (#18756)

DarkLight1337
authored and
amitm02
committed
[Bugfix] Mistral tool calling when content is list (#18729)

mgoin
authored and
amitm02
committed
[CI/Build] [TPU] Fix TPU CI exit code (#18282)

CAROLZXYZXY
authored and
amitm02
committed
[Neuron] Support quantization on neuron (#18283)

aws-satyajith
authored and
amitm02
committed
Support datasets in `vllm bench serve` and sync with benchmark_[serving,datasets].py (#18566)

mgoin
authored and
amitm02
committed
[Bugfix] Disable prefix caching by default for benchmark (#18771)

cascade812
authored and
amitm02
committed
[Build] Fixes for CMake install (#18570)

ProExpertProg
authored and
amitm02
committed
[Core] Improve Tensor serialisation (#18774)

lgeiger
authored and
amitm02
committed
[rocm] Fix wrong attention log (#18764)

fxmarty-amd
authored and
amitm02
committed
[Bugfix] Fix nomic max_model_len (#18755)

noooop
authored and
amitm02
committed
[Bugfix]: correctly propagate errors message caught at the chat_templating step to the client (#18769)

gcalmettes
authored and
amitm02
committed
[V1] fix torch profiling for V1 offline scenarios (#18445)

divakar-amd
authored and
amitm02
committed
[V1] [Bugfix] eagle bugfix and enable correct lm_head for multimodal (2) (#18781)

RonaldBXu
authored and
amitm02
committed
[Bugfix][FailingTest]Fix test_model_load_with_params.py (#18758)

rabi
authored and
amitm02
committed
[Deprecation] Require overriding `get_dummy_text` and `get_dummy_mm_data` (#18796)

DarkLight1337
authored and
amitm02
committed
[Misc][Tools][Benchmark] Add benchmark_serving supports for llama.cpp. (#18692)

Duyi-Wang
authored and
amitm02
committed
[Deprecation] Remove unused sync methods in `async_timeout` (#18792)

DarkLight1337
authored and
amitm02
committed
[Doc] Fix codeblocks formatting in LoRA adapters documentation (#18907)

Zerohertz
authored and
amitm02
committed
[Deprecation] Remove fallbacks for Embeddings API (#18795)

DarkLight1337
authored and
amitm02
committed
[Bugfix] Fix the failing gte embedding test (#18720)

Isotr0py
authored and
amitm02
committed
[CI] improve embed testing (#18747)

noooop
authored and
amitm02
committed
[Attention][V1] Toggle for v1 attention backend (#18275)

gshtras
authored and
amitm02
committed
Fix PiecewiseCompileInterpreter (#17338)

zou3519
authored and
amitm02
committed
[ROCm][V0][Attention] Revert to the previous FA triton kernel (#18226)

gshtras
authored and
amitm02
committed
[BugFix] FA2 MLA Accuracy Issue (#18807)

LucasWilkinson
authored and
amitm02
committed
[Deprecation] Disallow pos-args other than `model` when initializing `LLM` (#18802)

DarkLight1337
authored and
amitm02
committed
[Platform][Dist] Make torch distributed process group extendable (#18763)

MengqingCao
authored and
amitm02
committed
[Misc] Remove duplicate init for self.vllm_config (#18896)

googs1025
authored and
amitm02
committed
Enable Pydantic mypy checks and convert configs to Pydantic dataclasses (#17599)

hmellor
authored and
amitm02
committed
[V1] Allocate kv_cache with stride order for V1 (#18775)

NickLucche
authored and
amitm02
committed
[Frontend] add run batch to CLI (#18804)

reidliu41
authored and
amitm02
committed
[BugFix] Make DP work with connector-delayed new requests (#18559)

authored and
amitm02
committed
decrement server_load on listen for disconnect (#18784)

daniel-salib
authored and
amitm02
committed
[P/D] NixlConnector DP fixes (#18903)

wseaton
authored and
amitm02
committed
[Core] Add Lora Support to Beam Search (#18346)

alex-jw-brooks
authored and
amitm02
committed
Use standalone_compile by default in torch >= 2.8.0 (#18846)

zou3519
authored and
amitm02
committed
[Chore] update ty configuration (#18839)

aarnphm
authored and
amitm02
committed
[TPU] remove transpose ops in moe kernel (#18923)

yaochengji
authored and
amitm02
committed
[Misc] fix olmoe model layer can't laod in tp gt 1 (#18828)

lengrongfu
authored and
amitm02
committed
[Bugfix] Fix PP default fallback behavior for V1 (#18915)

mgoin
authored and
amitm02
committed
[V1][Metrics] Remove metrics that were deprecated in 0.8 (#18837)

markmc
authored and
amitm02
committed
[Misc] Update type annotation for rotary embedding `base` (#18914)

DarkLight1337
authored and
amitm02
committed
[Chore][Spec Decode] Update check NoneType instead of assigning variables (#18836)

aarnphm
authored and
amitm02
committed
[TPU][CI/CD] Clean up docker for TPU tests. (#18926)

CAROLZXYZXY
authored and
amitm02
committed
[Hardware][TPU][V1] Multi-LoRA Optimisations for the V1 TPU backend (#15655)

authored and
amitm02
committed
improve the robustness of parsing vlms config in AutoRound (#18894)

wenhuach21
authored and
amitm02
committed
Remove checks for `None` for fields which should never be `None` (#17985)

hmellor
authored and
amitm02
committed
[Bugfix] Consistent ascii handling in tool parsers (#18883)

chaunceyjiang
authored and
amitm02
committed
[Core] Enable CUDA graphs for DP + All2All kernels (#18724)

authored and
amitm02
committed
[Model] Use AutoWeightsLoader for mamba2 (#18918)

jinyouzhi
authored and
amitm02
committed
[Bugfix][ROCm] fix the power of 2 exception from triton_unified_attention.py when running llama4 models and unit test fix (#18100)

authored and
amitm02
committed
[docs] fix: fix markdown syntax (#18927)

eric-haibin-lin
authored and
amitm02
committed
Prevent the cross-encoder logic from being applied to classification tasks (#18838)

authored and
amitm02
committed
[ROCm] Remove unnecessary assertion of max_model_len in ROCM_AITER_MLA attention backend. (#18938)

vllmellm
authored and
amitm02
committed
Add ability to use CUDAGraphs with use_inductor=False (#17345)

zou3519
authored and
amitm02
committed
[Bugfix] Remove NVFP4 scales assertions to fix load_format=dummy (#18861)

mgoin
authored and
amitm02
committed
[Bugfix][TPU] fix moe custom kernel import (#18853)

yaochengji
authored and
amitm02
committed
[Deprecation] Remove mean pooling default for `Qwen2EmbeddingModel` (#18913)

DarkLight1337
authored and
amitm02
committed
[Doc][Neuron] Update documentation for Neuron (#18868)

elaineyz
authored and
amitm02
committed
[Misc]Fix benchmarks/README.md for speculative decoding (#18897)

rabi
authored and
amitm02
committed
Skip device and quant Pydantic validation to make plugin device work (#18843)

Yikun
authored and
amitm02
committed
[doc] add mkdocs doc (#18930)

reidliu41
authored and
amitm02
committed
Fixes a dead link in nightly benchmark readme (#18856)

nerdalert
authored and
amitm02
committed
[Model] Use in-place adds in SigLIP (#18922)

lgeiger
authored and
amitm02
committed
[Neuron] Add multi-LoRA support for Neuron. (#18284)

aws-satyajith
authored and
amitm02
committed
[Bugfix][Failing Test] Fix test_vllm_port.py (#18618)

rabi
authored and
amitm02
committed
[LoRA] Add LoRA support for InternVL (#18842)

jeejeelee
authored and
amitm02
committed
[Misc]Fix typo (#18947)

Always-Naive
authored and
amitm02
committed
[Doc] Remove redundant spaces from compatibility_matrix.md (#18891)

windsonsea
authored and
amitm02
committed
[Bugfix][TPU] Fix tpu model runner testcase failure (#18810)

CAROLZXYZXY
authored and
amitm02
committed
[doc] add CLI doc (#18871)

reidliu41
authored and
amitm02
committed
[CI/Build] remove regex from build dependencies (#18945)

authored and
amitm02
committed
[Bugfix] Fix misleading information in the documentation (#18845)

jeejeelee
authored and
amitm02
committed
[Feature] minicpm eagle support (#18943)

authored and
amitm02
committed
[Misc] Replace TODO in serving transcription (#18895)

NickLucche
authored and
amitm02
committed
[doc] show the count for fork and watch (#18950)

reidliu41
authored and
amitm02
committed
[Bugfix] Ensure tensors are contiguous during serialisation (#18860)

lgeiger
authored and
amitm02
committed
[Docs] Update SECURITY.md with link to our security guide (#18961)

russellb
authored and
amitm02
committed
[BugFix] Update pydantic to fix error on python 3.10 (#18852)

ProExpertProg
authored and
amitm02
committed
Improve "failed to get the hash of the compiled graph" error (#18956)

zou3519
authored and
amitm02
committed
Fix an error in dummy weight loading for quantization models (#18855)

Chenyaaang
authored and
amitm02
committed
[Perf] API-server scaleout with many-to-many server-engine comms (#17546)

njhill
authored and
amitm02
committed
Benchmark script for fp8 vs bf16 gemm (#17126)

mgoin
authored and
amitm02
committed
[VLM] Add PP support and fix GPTQ inference for Ovis models (#18958)

Isotr0py
authored and
amitm02
committed
[Misc] add group_size is -1 in awq quantization (#18910)

lengrongfu
authored and
amitm02
committed
Tool parser regex timeout handling (#18960)

wseaton
authored and
amitm02
committed
[Docs] Correct multiprocessing design doc (#18964)

lgeiger
authored and
amitm02
committed
create util function for batched arange (#18937)

yuguo68
authored and
amitm02
committed
[Frontend] Add rerank support to run_batch endpoint (#16278)

pooyadavoodi
authored and
amitm02
committed
[Misc] Fix estimated max model len msg (#18966)

sarckk
authored and
amitm02
committed
[Bugfix]: Fix the incompatibility issue with Structured Outputs when Thinking is disabled (#18879)

chaunceyjiang
authored and
amitm02
committed
fix security issue of logging llm output (#18980)

authored and
amitm02
committed
[Neuron] Add Multi-Modal model support for Neuron (#18921)

authored and
amitm02
committed
[doc] fix the list rendering issue - security.md (#18982)

reidliu41
authored and
amitm02
committed
[BugFix] Pydantic part 2 (#18911)

ProExpertProg
authored and
amitm02
committed
[FEAT][ROCm] Add AITER grouped topk for DeepSeekV2 (#18825)

vllmellm
authored and
amitm02
committed
[Bugfix] Fix for issue 17396 (#18773)

frreiss
authored and
amitm02
committed
[ROCm][Kernel] Add gfx950 support for skinny gemms (#18010)

charlifu
authored and
amitm02
committed
[P/D] NixlConnector use cache device index for memory registration (#18969)

ptarasiewiczNV
authored and
amitm02
committed
[BugFix] Fix multi-node offline data-parallel (#18981)

authored and
amitm02
committed
[Misc] add return token strs for tokenize (#18941)

reidliu41
authored and
amitm02
committed
[Misc][Benchmark] Add support for CustomDataset (#18511)

ekagra-ranjan
authored and
amitm02
committed
[Bugfix] Fix EAGLE3 broken logits (#18909)

benchislett
authored and
amitm02
committed
[Core] Rework dtype resolution (#18751)

DarkLight1337
authored and
amitm02
committed
[LoRA] Support dynamically initialize `packed_modules_mapping` for VLM with arbitrary components (#18987)

Isotr0py
authored and
amitm02
committed
[doc] small fix - mkdocs (#18996)

reidliu41
authored and
amitm02
committed
Let max_num_batched_tokens use human_readable_int for large numbers (#18968)

mgoin
authored and
amitm02
committed
[BugFix] fix data parallel construct ipv6 url addres (#18991)

lengrongfu
authored and
amitm02
committed
[BugFix] Fix incorrect metrics shutdown error log message (#18992)

njhill
authored and
amitm02
committed
[doc] wrong output (#19000)

reidliu41
authored and
amitm02
committed
Update request.py
amitm02
committed
Update request.py
amitm02
committed
[Misc][Tools][Benchmark] Add benchmark_serving supports for llama.cpp. (#18692)

Duyi-Wang
authored and
amitm02
committed
[Doc] Fix codeblocks formatting in LoRA adapters documentation (#18907)

Zerohertz
authored and
amitm02
committed
[Bugfix] Fix the failing gte embedding test (#18720)

Isotr0py
authored and
amitm02
committed
[Attention][V1] Toggle for v1 attention backend (#18275)

gshtras
authored and
amitm02
committed
[ROCm][V0][Attention] Revert to the previous FA triton kernel (#18226)

gshtras
authored and
amitm02
committed
[Deprecation] Disallow pos-args other than `model` when initializing `LLM` (#18802)

DarkLight1337
authored and
amitm02
committed
[Misc] Remove duplicate init for self.vllm_config (#18896)

googs1025
authored and
amitm02
committed
[V1] Allocate kv_cache with stride order for V1 (#18775)

NickLucche
authored and
amitm02
committed
[BugFix] Make DP work with connector-delayed new requests (#18559)

authored and
amitm02
committed
[P/D] NixlConnector DP fixes (#18903)

wseaton
authored and
amitm02
committed
Use standalone_compile by default in torch >= 2.8.0 (#18846)

zou3519
authored and
amitm02
committed
[TPU] remove transpose ops in moe kernel (#18923)

yaochengji
authored and
amitm02
committed
[Bugfix] Fix PP default fallback behavior for V1 (#18915)

mgoin
authored and
amitm02
committed
[Misc] Update type annotation for rotary embedding `base` (#18914)

DarkLight1337
authored and
amitm02
committed
[TPU][CI/CD] Clean up docker for TPU tests. (#18926)

CAROLZXYZXY
authored and
amitm02
committed
improve the robustness of parsing vlms config in AutoRound (#18894)

wenhuach21
authored and
amitm02
committed
[Bugfix] Consistent ascii handling in tool parsers (#18883)

chaunceyjiang
authored and
amitm02
committed
[Model] Use AutoWeightsLoader for mamba2 (#18918)

jinyouzhi
authored and
amitm02
committed
[docs] fix: fix markdown syntax (#18927)

eric-haibin-lin
authored and
amitm02
committed
[ROCm] Remove unnecessary assertion of max_model_len in ROCM_AITER_MLA attention backend. (#18938)

vllmellm
authored and
amitm02
committed
[Bugfix] Remove NVFP4 scales assertions to fix load_format=dummy (#18861)

mgoin
authored and
amitm02
committed
[Deprecation] Remove mean pooling default for `Qwen2EmbeddingModel` (#18913)

DarkLight1337
authored and
amitm02
committed
[Misc]Fix benchmarks/README.md for speculative decoding (#18897)

rabi
authored and
amitm02
committed
[doc] add mkdocs doc (#18930)

reidliu41
authored and
amitm02
committed
[Model] Use in-place adds in SigLIP (#18922)

lgeiger
authored and
amitm02
committed
[Bugfix][Failing Test] Fix test_vllm_port.py (#18618)

rabi
authored and
amitm02
committed
[Misc]Fix typo (#18947)

Always-Naive
authored and
amitm02
committed
[Bugfix][TPU] Fix tpu model runner testcase failure (#18810)

CAROLZXYZXY
authored and
amitm02
committed
[CI/Build] remove regex from build dependencies (#18945)

authored and
amitm02
committed
[Feature] minicpm eagle support (#18943)

authored and
amitm02
committed
[doc] show the count for fork and watch (#18950)

reidliu41
authored and
amitm02
committed
[Docs] Update SECURITY.md with link to our security guide (#18961)

russellb
authored and
amitm02
committed
Improve "failed to get the hash of the compiled graph" error (#18956)

zou3519
authored and
amitm02
committed
[Perf] API-server scaleout with many-to-many server-engine comms (#17546)

njhill
authored and
amitm02
committed
Benchmark script for fp8 vs bf16 gemm (#17126)

mgoin
authored and
amitm02
committed
[VLM] Add PP support and fix GPTQ inference for Ovis models (#18958)

Isotr0py
authored and
amitm02
committed
[Misc] add group_size is -1 in awq quantization (#18910)

lengrongfu
authored and
amitm02
committed
Tool parser regex timeout handling (#18960)

wseaton
authored and
amitm02
committed
[Docs] Correct multiprocessing design doc (#18964)

lgeiger
authored and
amitm02
committed
create util function for batched arange (#18937)

yuguo68
authored and
amitm02
committed
[Frontend] Add rerank support to run_batch endpoint (#16278)

pooyadavoodi
authored and
amitm02
committed
[Misc] Fix estimated max model len msg (#18966)

sarckk
authored and
amitm02
committed
[Bugfix]: Fix the incompatibility issue with Structured Outputs when Thinking is disabled (#18879)

chaunceyjiang
authored and
amitm02
committed
fix security issue of logging llm output (#18980)

authored and
amitm02
committed
[Neuron] Add Multi-Modal model support for Neuron (#18921)

authored and
amitm02
committed
[doc] fix the list rendering issue - security.md (#18982)

reidliu41
authored and
amitm02
committed
[BugFix] Pydantic part 2 (#18911)

ProExpertProg
authored and
amitm02
committed
[FEAT][ROCm] Add AITER grouped topk for DeepSeekV2 (#18825)

vllmellm
authored and
amitm02
committed
[Bugfix] Fix for issue 17396 (#18773)

frreiss
authored and
amitm02
committed
[ROCm][Kernel] Add gfx950 support for skinny gemms (#18010)

charlifu
authored and
amitm02
committed
[P/D] NixlConnector use cache device index for memory registration (#18969)

ptarasiewiczNV
authored and
amitm02
committed
[BugFix] Fix multi-node offline data-parallel (#18981)

authored and
amitm02
committed
[Misc] add return token strs for tokenize (#18941)

reidliu41
authored and
amitm02
committed
[Misc][Benchmark] Add support for CustomDataset (#18511)

ekagra-ranjan
authored and
amitm02
committed
[Bugfix] Fix EAGLE3 broken logits (#18909)

benchislett
authored and
amitm02
committed
[Core] Rework dtype resolution (#18751)

DarkLight1337
authored and
amitm02
committed
[LoRA] Support dynamically initialize `packed_modules_mapping` for VLM with arbitrary components (#18987)

Isotr0py
authored and
amitm02
committed
[doc] small fix - mkdocs (#18996)

reidliu41
authored and
amitm02
committed
Let max_num_batched_tokens use human_readable_int for large numbers (#18968)

mgoin
authored and
amitm02
committed
[BugFix] fix data parallel construct ipv6 url addres (#18991)

lengrongfu
authored and
amitm02
committed
[BugFix] Fix incorrect metrics shutdown error log message (#18992)

njhill
authored and
amitm02
committed
[doc] wrong output (#19000)

reidliu41
authored and
amitm02
committed
Update request.py
amitm02
committed
Update request.py
amitm02
committed
Merge branch 'feat/v1-priority-scheduling' of https://github.com/amitm02/vllm into feat/v1-priority-scheduling
amitm02
committed
Merge remote-tracking branch 'upstream/main' into feat/v1-priority-scheduling
amitm02
committed
minor
amitm02
committed
line too long
amitm02
committed
line too long
amitm02
committed

Commits on Jun 3, 2025

Merge remote-tracking branch 'upstream/main' into feat/v1-priority-scheduling
amitm02
committed