# vLLM v0.17.0 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.17.0
- **Released:** Mar 07, 2026
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.17.0/

## Executive Summary (TL;DR)
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:

## Breaking Changes & Deprecations
- PyTorch 2.10 Upgrade**: This release upgrades to **PyTorch 2.10.0**, which is a breaking change for environment dependencies.
- Breaking**: KV load failure policy default changed from "recompute" to "fail" (#34896).
- PyTorch 2.10.0 upgrade** — breaking change requiring environment updates. ROCm torch also updated to official 2.10 release (#34387).

## New Features
- Structured output JSON feature validation (#33233).
- mooncake-transfer-engine added to kv_connectors requirements (#34826).
- amd-quark package added for ROCm (#35658).

## Improvements
- **FlashAttention 4 Integration**: vLLM now supports the **FlashAttention 4** backend (#32974), bringing next-generation attention performance.
- **New `--performance-mode` Flag**: A new `--performance-mode {balanced, interactivity, throughput}` flag (#34936) simplifies performance tuning for common deployment scenarios.
- **Performance**: Qwen3.5 GDN projector fusion (#34697), FlashInfer cuDNN backend for Qwen3 VL ViT (#34580), Step3.5-Flash NVFP4 (#34478), Qwen3MoE tuned configs for H200 (#35457).
- **NVIDIA**: SM100 FMHA FP8 prefill for MLA (#31195), SM100 MXFP8 blockscaled grouped MM and quant kernels (#34448), SM100 Oink RMSNorm path (#31828), SM120 FP8 GEMM optimization (#34424), FlashInfer DeepGEMM swapAB on SM90 by default (#34924), DeepSeek R1 BF16 min latency QKV GEMM 0.5% E2E speedup (#34758), Cublas BF16 gate with FP32 output (#35121), FlashInfer All Reduce default to TRTLLM backend (#35793).
- **AMD ROCm**: AITER fused RoPE+KVCache (#33443), MXFP4 MoE weight pre-shuffling on gfx950 (#34192), bitsandbytes quantization (#34688), CK backend for MoE quantization (#34301), dynamic MXFP4 for DeepSeek V2 (#34157), GPT-OSS Quark format (#29008), GPT-OSS WMXFP4_AFP8 static scales (#30357), encoder/encoder-decoder on AITER (#35334), device capability derivation without CUDA init (#35069), `aiter` package renamed to `amd-aiter` (#35198).
- **Intel XPU**: CUDA graph support (#34482), GPUDirect RDMA via NIXL (#35270), TORCH_SDPA/TRITON_ATTN as ViT backend (#35010), vllm-xpu-kernels v0.1.3 (#35984).

## Bug Fixes
- **Anthropic API Compatibility**: Added support for **Anthropic thinking blocks** (#33671), **`count_tokens` API** (#35588), `tool_choice=none` (#35835), and streaming/image handling fixes.
- **CPU release supports AVX2, AVX-512, VNNI, AVX512BF16, and AMX** (#35466). The multi-ISA CPU dispatcher was originally implemented by @MekayelAnik (https://github.com/dtrifiro/vllm/pull/9, merged December 22, 2025) in collaboration with Willy Hardy, and later reimplemented in C++ in #35466.
- **Fixes**: DeepSeek-VL V2 simplified loading (#35203), Qwen3/Qwen3.5 reasoning parser (#34779), Qwen2.5-Omni/Qwen3-Omni mixed-modality (#35368), Ernie4.5-VL garbled output (#35587), Qwen-VL tokenizer (#36140), Qwen-Omni audio cache (#35994), Nemotron-3-Nano NVFP4 accuracy with TP>1 (#34476).
- **Fixes**: allreduce_rms_fusion disabled by default with PP > 1 (#35424), DCP + FA3 crash (#35082), prefix caching for Mamba "all" mode (#34874), num_active_loras fix (#34119), async TP reduce-scatter reduction fix (#33088).
- Cross-node data parallelism message queue fix (#35429).
- Fix routing for models without expert groups (MiniMax-M2.1) (#34673).

## Upgrade Command
```bash
python -m pip install 'vllm==0.17.0'
```