# vLLM v0.15.1 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.15.1
- **Released:** Feb 04, 2026
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.15.1/

## Executive Summary (TL;DR)
vLLM v0.15.1. Source highlight: v0.15.1 is a patch release with security fixes, RTX Blackwell GPU fixes support, and bug fixes.

## Improvements
- **MoE forward pass**: Optimized by caching layer name computation (#33184)

## Bug Fixes
- **RTX Blackwell (SM120)**: Fixed NVFP4 MoE kernel support for RTX Blackwell workstation GPUs. Previously, NVFP4 MoE models would fail to load on these GPUs (#33417)
- **FP8 kernel selection**: Fixed FP8 CUTLASS group GEMM to properly fall back to Triton kernels on SM120 GPUs (#33285)
- **Qwen3-VL-Reranker**: Fixed model loading (#33298)
- **Whisper**: Fixed FlashAttention2 with full CUDA graphs (#33360)
- **torch.compile cold-start**: Fixed regression that increased cold-start compilation time (Llama3-70B: ~88s → ~22s) (#33441)
- Fixed prefix cache hit rate of 0% with GPT-OSS style hybrid attention models (#33524)

## Upgrade Command
```bash
python -m pip install 'vllm==0.15.1'
```