# vLLM v0.9.0 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.9.0
- **Released:** May 15, 2025
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.9.0/

## Executive Summary (TL;DR)
vLLM v0.9.0. Source highlight: vLLM has upgraded to PyTorch 2.7!  (#16859) This is a breaking change for environment dependency.

## Breaking Changes & Deprecations
- vLLM has upgraded to PyTorch 2.7!  (#16859) This is a breaking change for environment dependency.
- [Bugfix][ROCm] Restrict ray version due to a breaking release by @gshtras in https://github.com/vllm-project/vllm/pull/17480

## New Features
- Add ep group and all2all interface (#18077)
- Add `VLLM_ALLOW_INSECURE_SERIALIZATION` env var (#17490)
- CLI: `deprecated=True` (#17426)
- Frontend: progress bar for adding requests (#17525), `chat_template_kwargs` in `LLM.chat` (#17356), `/classify` endpoint (#17032), truncation control for embedding models (#14776), `cached_tokens` in response usage (#18149)
- LoRA: default local directory LoRA resolver plugin. (#16855)
- Metrics: kv event publishing (#16750), API for accessing in-memory Prometheus metrics (#17010)

## Improvements
- Enhanced NVIDIA Blackwell support. vLLM now ships with initial set of optimized kernels on NVIDIA Blackwell with both attention and mlp.
- You can use our docker image or install FlashInfer nightly wheel `pip install https://download.pytorch.org/whl/cu128/flashinfer/flashinfer_python-0.2.5%2Bcu128torch2.7-cp38-abi3-linux_x86_64.whl` then set `VLLM_ATTENTION_BACKEND=FLASHINFER` for better performance.
- Support MiMo-7B (#17433), MiniMax-VL-01 (#16328), Ovis 1.6 (#17861), Ovis 2 (#15826), GraniteMoeHybrid 4.0 (#17497), FalconH1\* (#18406), LlamaGuard4 (#17315)
- Please install the development version of `transformers` (from source) to use Falcon-H1.
- Embedding models: nomic-embed-text-v2-moe (#17785), new class of gte models (#17986)
- Progress in Hybrid Memory Allocator (#17394, #17479, #17474, #17483, #17193, #17946, #17945, #17999, #18001, #18593)

## Bug Fixes
- Fix image hash collision in certain edge cases (#17378)
- [Bugfix] Fix `numel()` downcast in fused_layernorm_dynamic_per_token_quant.cu
- [Bugfix] Add contiguous call inside rope kernel wrapper
- [Fix] Documentation spacing in compilation config help text
- [Build][Bugfix] Restrict setuptools version to <80
- [Doc] Fix QWen3MOE info

## Upgrade Command
```bash
python -m pip install 'vllm==0.9.0'
```