# vLLM v0.26.0 — Release & Upgrade Intelligence
- **Product:** vLLM (https://vllm.ai)
- **Version:** v0.26.0
- **Released:** Jul 27, 2026
- **Impact Score:** 5/10
- **Breaking Changes:** Not verified
- **Canonical Analysis:** https://changelogs.us/vllm/updates/v0.26.0/

## Executive Summary (TL;DR)
vLLM v0.26.0. Source highlight: vLLM v0.26.0 Release Notes

## New Features
- Add resource-bounds validation to derender endpoints (#47260); sanitize server file paths from validation error responses (#46415); bound the completion prompt list to prevent unbounded engine fan-out (#47845); guard lm-format-enforcer regex compilation with a timeout (#47595).

## Improvements
- **DeepSeek-V4 performance push** across vendors: a specialized routing kernel (2.94% E2E TPOT, #48660), `fused_topk_bias` (1.5–2x kernel, #47463), and redundant repeat/copy removal (1.8% E2E TPOT, #48137), plus ROCm two-stage compressor for HCA prefill (#47718), sparse decode/prefill optimizations (#48519, #48788, #46275), and DSpark speculative decoding on AMD (#47419) and XPU (#47677).
- **fp32 `lm_head` for generation models via `head_dtype`** (#48390), extended to the LoRA path (#48525) and given a ROCm `torch.mm` fast path (#48688), improving accuracy for generation heads.
- **Flexible attention backends**: the attention backend can now be selected per KV-cache group (#48012), and sliding-window support is now an explicit backend capability (#48011) — improving support for hybrid models.
- fp32 `lm_head` for generation models via `head_dtype` (#48390); lower memory for capturing large CUDA graph sizes (#48483); opt-in persistence and reuse of the memory-profiling result across boots (#47388); improved InstantTensor loading (#46868).
- DeepSeek-V4: specialized routing kernel (2.94% E2E TPOT, #48660), `fused_topk_bias` 1.5–2x (#47463), redundant repeat/copy removal (1.8% TPOT, #48137).
- MoE router GEMMs: BF16x3 router GEMM (#47973), FP32 router GEMV (#48335), generic CuteDSL LL BF16 router GEMM (#42562); TRTLLM BF16 MoE modular kernel (#45182); write FlashInfer combine into final output (#47156).

## Bug Fixes
- Fixes: correct pooling scores for chunked prefill under `torch.compile` (#48901).
- Fixes: host memory leak from undrained `new_block_ids` (#44490), DSv3.2 + MTP + sequence-parallel accuracy (#48036).
- Stability/correctness: handle grammar compilation failures without crashing the engine (#47312), fix logprobs token-string collision from SentencePiece spaces (#48674).
- Fix a concurrent sparse-invariant race that bypassed CVE remediation (#48583).

## Upgrade Command
```bash
python -m pip install 'vllm==0.26.0'
```