# vLLM v0.16.0 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.16.0
- **Released:** Feb 25, 2026
- **Impact Score:** 7/10
- **Breaking Changes:** Source flagged
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.16.0/

## Executive Summary (TL;DR)
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.

## Breaking Changes & Deprecations
- Deprecation & Breaking Changes * Removed BitBlas quantization (#32683) and Marlin 24 (#32688). * Removed deprecated `reasoning_content` message field (#33402).

## New Features
- Features: run_batch transcription/translation support (#33934), /server_info collect_env (#33246), OTEL tracing during model loading (#31162), clear MM and encoder cache (#33452), HF Hub LoRA resolver (#20320).

## Improvements
- **Async scheduling + Pipeline Parallelism** is now fully supported, delivering **30.8% E2E throughput improvement** and **31.8% TPOT improvement** (#32618).
- **RLHF workflow improvements**: Native NCCL-based weight syncing API (#31943), layerwise weight reloading for QeRL (#32133), and engine pause/resume with request preservation (#32351).
- **Intel XPU**: Platform overhaul - deprecated IPEX, switched to vllm-xpu-kernels (#33379). New: unquantized MoE (#33659), MXFP4 MoE (#33679), WNA16 kernel (#33973), scaled_mm kernel (#34117), FP8 MoE (#34202).
- **IBM Z**: BF16 kernel type for s390x (#33788).
- **Performance**: Chat completion streaming optimization (#33782), ORJSONResponse for faster API responses (#33548), MoE permute optimization for CUTLASS FP8 (#32892), shared/routed overlap for latent MoE on Nemotron-H (#32790), FlashInfer autotune control flag (#34006).

## Bug Fixes
- Features: Qwen3-Omni transcription (#29828), Mistral Large 3 with FlashInfer MoE (#33174), LFM2 SigLIP2 intermediate encoder layers (#33370), Qwen3-Omni/GLM-4.xV MRoPE positioning fixes (#33010, #33039), embedding input for disabled modalities (#32493).
- Performance: GLM-4.7-GPTQ decode and MTP acceptance rate regression fix (#33771), DeepSeek V3.2 fast detokenization (#33855), DeepSeek V3.2 tokenizer fix (#33832), GLM-5 MTP accuracy fix (#34385).
- Async scheduling + Pipeline Parallelism: Full support with 30.8% throughput improvement (#32618), optimized spec decode + async scheduling with 1.5% throughput improvement (#33612), deadlock fix for torchrun PP broadcast (#33701).
- Speculative decoding: Unified Parallel Drafting (#32887), structured output support (#33374), penalty application in MRV2 (#33251), skip softmax for all-greedy rejection sampling (#32852), correctness fix for spec tokens with prefill chunks (#33652).
- Performance: Grammar bitmask H2D copy on separate stream (#33059), zero-copy GQA for multimodal and CPU (#33732), early-reject oversized MM requests (#33502), CPU memory leak fix from Request reference cycle in prefix caching (#34183).
- **NVIDIA**: FlashInfer TRTLLM BF16 MoE integration (#32954), SM100 INT4 W4A16 kernel (#32437), SM121 (DGX Spark) CUTLASS support (#33517), MNNVL protocol for GB series (#33540), FlashInfer MLA concat optimization (#31171), GDN attention layout optimization (#33291), DeepGEMM FP8 MLA performance (#33568), wvSplitK_fp8 performance (#33527, #33493), B200 MoE configs for Nemotron Nano (#32804), Super B200 TP2 (#33510), GLM 4.6 (#32958), Mamba selective scan tuning for B200 (#32873). Fix: DeepSeek R1

## Upgrade Command
```bash
python -m pip install 'vllm==0.16.0'
```