Highlights This release features 584 commits from 270 contributors (76 new)! Kimi-K3 performance push: a major optimization effort for Kimi-K3 across the stack — Decode Context Parallel (DCP) support (#50484), fused FlashKDA decode and pref
vLLM breaking changes
The 16 releases whose own notes call out a breaking change, out of 60 the archive holds. Scattered one per release page, this is a run nobody can see; gathered, it is the shape of how often vLLM asks you to change your code.
- Tagged breaking
- 16of 60 releases
- Roughly
- 1 in 4releases held here
- Most recent
- 26 Aug 20262 days ago
- Across
- 23 monthsback to 25 Sep 2024
How this is decided
A release appears here because the word breaking is in the notes its publisher wrote, not because anyone here read the diff. That cuts both ways. A release that changed an API without using the word is missing from this page, and a release that used the word about something minor is on it. It is the publisher’s own signal, reported rather than judged, which is the same rule the security page follows.
Newest first
All vLLM releasesvLLM v0.20.0 Highlights This release features 752 commits from 320 contributors (123 new)! DeepSeek V4: Initial DeepSeek V4 support landed (#40860), with DSML token-leakage fix in DSV4/3.2 (#40806), DSA + MTP IMA fix (#40772), and a silu cl
vLLM v0.18.0 Known issues Degraded accuracy when serving Qwen3.5 with FP8 KV cache on B200 (#37618) If you previously ran into CUBLAS_STATUS_INVALID_VALUE and had to use a workaround in v0.17.0, you can reinstall torch 2.10.0. PyTorch publi
vLLM v0.17.0 Known Issue: If you are on CUDA 12.9+ and encounter a CUBLAS_STATUS_INVALID_VALUE error, this is caused by a CUDA library mismatch. To resolve, try one of the following: Remove the path to system CUDA shared library files (e.g.
vLLM v0.16.0 Please note that this release was branch cut on Feb 8, so any features added to vLLM after that date is not included. Highlights This release features 440 commits from 203 contributors (7 new)! Async scheduling + Pipeline Paral
Highlights This release features 335 commits from 158 contributors (39 new)! Model Support New architectures: Kimi-K2.5 (#33131), Molmo2 (#30997), Step3vl 10B (#32329), Step1 (#32511), GLM-Lite (#31386), Eagle2.5-8B VLM (#32456). LoRA expan
Highlights This release features approximately 660 commits from 251 contributors (86 new contributors). Breaking Changes: Async scheduling is now enabled by default - Users who experience issues can disable with --no-async-scheduling. Exclu
vLLM v0.13.0 Release Notes Highlights Highlights This release features 442 commits from 207 contributors (61 new contributors)! Breaking Changes: This release includes deprecation removals, PassConfig flag renames, and attention configurati
vLLM v0.12.0 Release Notes Highlights Highlights This release features 474 commits from 213 contributors (57 new)！ Breaking Changes: This release includes PyTorch 2.9.0 upgrade (CUDA 12.9), V0 deprecations including xformers backend, and sc
Highlights This release contains 740 commits from 266 contributors (97 new)! Breaking Changes: This release includes PyTorch 2.8.0 upgrade, V0 deprecations, and API changes - please review the changelog carefully. aarch64 support: This rele
Highlights v0.10.1 release includes 727 commits, 245 committers (105 new contributors). NOTE: This release deprecates V0 FA3 support and as a result FP8 kv-cache in V0 may have issues Model Support New model families: GPT-OSS with comprehen
Highlights This release features 649 commits, from 215 contributors (82 new contributors!) vLLM has upgraded to PyTorch 2.7! (#16859) This is a breaking change for environment dependency. The default wheel has been upgraded from CUDA 12.4 t
This release contains 310 commits from 143 contributors (55 new contributors!). Highlights This release features important multi-modal bug fixes, day 0 support for Qwen3, and xgrammar's structure tag feature for tool calling. Model Support
featured 523 commits from 166 total contributors (68 new contributors)! Highlights V1 We have now enabled V1 engine by default (#13726) for supported use cases. Please refer to V1 user guide for more detail. We expect better performance for
Highlights Support Deepseek V3 (#11523, #11502) model. On 8xH200s or MI300x: vllm serve deepseek-ai/DeepSeek-V3 --tensor-parallel-size 8 --trust-remote-code --max-model-len 8192. The context length can be increased to about 32K beyond runni
Highlights Model Support Support Llama 3.2 models (#8811, #8822) vllm serve meta-llama/Llama-3.2-11B-Vision-Instruct --enforce-eager --max-num-seqs 16 Beam search have been soft deprecated. We are moving towards a version of beam search tha