91 tracked entries. Older history is imported progressively; see source coverage.
vLLM v0.32.0 introduces Model Runner V2 as the default, integrates comprehensive structured-output support, and enforces significant dependency and configuration breaking changes. Users must prepare for Python 3.11 requirements, updated CORS defaults, and API shifts in reasoning streaming.
vLLM v0.31.0 introduces major architectural performance additions for DeepSeek-V4.1-Flash alongside strict security gating for multimodal request arguments and configuration parameter renames. Upgrading teams must review breaking changes around runtime parameters, removed backends, and modified quantization APIs.
vLLM v0.30.0. Source highlight: This release features 762 commits from 315 contributors (104 new)!
vLLM v0.29.0. Source highlight: This release features 594 commits from 277 contributors (91 new)!
vLLM v0.28.0. Source highlight: This release features 584 commits from 270 contributors (76 new)!
vLLM v0.27.1 is published, but the available notes are too limited to assess upgrade impact. Review the official source before upgrading.
vLLM v0.27.0. Source highlight: vLLM v0.27.0 Release Notes
vLLM v0.26.0. Source highlight: vLLM v0.26.0 Release Notes
vLLM v0.25.1. Source highlight: This release features 2 commits from 2 contributors (1 new)!
vLLM v0.25.0. Source highlight: vLLM v0.25.0 Release Notes
vLLM v0.24.0. Source highlight: vLLM v0.24.0 Release Notes
vLLM v0.23.0. Source highlight: vLLM v0.23.0 Release Notes
vLLM v0.22.1. Source highlight: This release features 8 commits from 6 contributors (1 new)!
vLLM v0.22.0. Source highlight: This release features 459 commits from 230 contributors (63 new)!
vLLM v0.21.0. Source highlight: This release features 367 commits from 202 contributors (49 new)!
vLLM v0.20.2. Source highlight: This release features 6 commits from 6 contributors (0 new)!
vLLM v0.20.1. Source highlight: This is a patch release on top of `v0.20.0` primarily focused on **DeepSeek V4 stabilization and performance improvements**, along with several important bug fixes.
vLLM v0.20.0. Source highlight: This release features 752 commits from 320 contributors (123 new)!
vLLM v0.19.1. Source highlight: This is a patch release on top of `v0.19.0` with Transformers v5.5.3 upgrade and bug fixes for Gemma4:
vLLM v0.19.0. Source highlight: This release features 448 commits from 197 contributors (54 new)!
vLLM v0.18.1. Source highlight: This is a patch release on top of v0.18.0 to address a few issues:
vLLM v0.18.0. Source highlight: Degraded accuracy when serving Qwen3.5 with FP8 KV cache on B200 (#37618)
vLLM v0.17.1. Source highlight: This is a patch release on top of `v0.17.0` to address a few issues:
vLLM v0.17.0. Source highlight: **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:
vLLM v0.16.0. Source highlight: Please note that this release was branch cut on Feb 8, so any features added to vLLM after that date is not included.
vLLM v0.15.1. Source highlight: v0.15.1 is a patch release with security fixes, RTX Blackwell GPU fixes support, and bug fixes.
vLLM v0.15.0. Source highlight: This release features 335 commits from 158 contributors (39 new)!
vLLM v0.14.1 is published, but the available notes are too limited to assess upgrade impact. Review the official source before upgrading.
vLLM v0.14.0. Source highlight: **Breaking Changes:**
vLLM v0.13.0. Source highlight: vLLM v0.13.0 Release Notes Highlights
vLLM v0.12.0. Source highlight: vLLM v0.12.0 Release Notes Highlights
vLLM v0.11.2. Source highlight: This release includes 4 bug fixes on top of `v0.11.1`:
vLLM v0.11.1. Source highlight: Key changes include:
vLLM v0.11.0. Source highlight: This release completes the removal of V0 engine. V0 engine code including AsyncLLMEngine, LLMEngine, MQLLMEngine, all attention backends, and related components have been removed. **V1 is the only engine in the codebase now.**
vLLM v0.10.2. Source highlight: This release contains 740 commits from 266 contributors (97 new)!
vLLM v0.10.1.1. Source highlight: This is a critical bugfix and security release:
vLLM v0.10.1. Source highlight: **NOTE: This release deprecates V0 FA3 support and as a result FP8 kv-cache in V0 may have issues**
vLLM v0.10.0. Source highlight: v0.10.0 release includes 308 commits, 168 contributors (62 new!).
vLLM v0.9.2. Source highlight: This release contains 452 commits from 167 contributors (31 new!)
vLLM v0.9.1. Source highlight: Progress in large scale serving
vLLM v0.9.0.1. Source highlight: This patch release contains important bugfix for DeepSeek family of models on NVIDIA Ampere and below (#18807)
vLLM v0.9.0. Source highlight: vLLM has upgraded to PyTorch 2.7! (#16859) This is a breaking change for environment dependency.
vLLM v0.8.5.post1. Source highlight: This post release contains two bug fix for memory leak and model accuracy
vLLM v0.8.5. Source highlight: This release features important multi-modal bug fixes, day 0 support for Qwen3, and xgrammar's structure tag feature for tool calling.
vLLM v0.8.4. Source highlight: This release includes important accuracy fixes for Llama4 models, if you are using it, we highly recommend you to update.
vLLM v0.8.3. Source highlight: We are excited to announce Day 0 Support for Llama 4 Scout and Maverick (#16104). Please see our blog for detailed user guide.
vLLM v0.8.2. Source highlight: This release contains important bug fix for the V1 engine's memory usage. We highly recommend you upgrading!
vLLM v0.8.1. Source highlight: This release contains important bug fixes for v0.8.0. We highly recommend upgrading!
vLLM v0.8.0. Source highlight: 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 supported scenarios. If you'd like to disable V1 mode, please specify the environment variable `VLLM_USE_V1=0`, and send us a GitHub issue sharin
vLLM v0.7.3. Source highlight: Deepseek enhancements: