# vLLM v0.27.0

- Product: vLLM (https://whatsnew.fyi/product/vllm)
- Vendor: vLLM
- Date: 2026-08-10
- Version: v0.27.0
- Original notes: https://github.com/vllm-project/vllm/releases/tag/v0.27.0
- Permalink: https://whatsnew.fyi/product/vllm/releases/v0.27.0

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **added** — Kimi K3 support with core model files, kernels, Python and Rust frontends, AttnRes kernels, DeepGEMM support, compressed-tensors quantized checkpoints, DSpark AR fusion, and optional shared expert sharding
- **added** — Qwen3.5 text-only dense and MoE models with EVS video token pruning
- **added** — K-EXAONE-2.0-750B-A37B model support
- **added** — VaultGemma support via the Transformers modeling backend
- **added** — jina-embeddings-v5-text-nano model with EuroBERT encoder backbone
- **changed** — Upgrade PyTorch to 2.13.0 along with torchvision 0.28.0 and Triton 3.7.1
- **changed** — XPU and CPU backends upgraded to PyTorch 2.13
- **added** — FlashAttention 4 integration on SM100 with FP8 KV cache support and headdim-256 support
- **added** — New JIT warmup infrastructure and runner-owned Triton kernel warmup to remove first-request compilation stalls
- **added** — DeepSeek-V4 sequence parallelism
- **changed** — DeepSeek-V4 kernel improvements including skipping empty c128 launches, workspace reuse, and removal of redundant full kernel
- **changed** — DeepSeek-V4 optimizations achieving 3.4% E2E TTFT improvement from skipping unneeded topk and router operations
- **changed** — DeepSeek-V4 optimizations achieving 3.9% E2E TTFT improvement from workspace reuse
- **added** — Compact MXFP4 indexer KV cache for DeepSeek-V4
- **changed** — Model Runner V2 expanded to encoder-only attention, sequence pooling for embedding and classification, encoder token classification and embedding, and BGE-M3 pooling
- **added** — Model Runner V2 multimodal support on CPU
- **added** — Multi-layer MTP speculator in Model Runner V2
- **added** — Fault tolerance framework for DP+EP external load-balancer deployments
- **added** — Async preparation for elastic EP scaling
- **added** — NIXL P/D support for hybrid MLA+SSM models and heterogeneous P/D block sizes for hybrid models
- **added** — Rust frontend gRPC control plane with engine-aware health reporting, abort control, server and model discovery, and KV event source discovery

#### vLLM v0.27.0 Release Notes

##### Highlights

This release features 561 commits from 242 contributors (64 new)!

* **Kimi K3 support** with a full stack landing in one release: core model files and kernels (#50089, #50000), Python (#50093) and Rust (#50104) frontends, AttnRes kernels (#50090), DeepGEMM support (#50458), compressed-tensors quantized checkpoints (#50500), DSpark AR fusion (#50242), and an option to shard the shared expert instead of replicating it (#50656).
* **More new models**: Qwen3.5 text-only dense and MoE models (#50210) with EVS video token pruning (#48912), K-EXAONE-2.0-750B-A37B (#50524), VaultGemma via the Transformers modeling backend (#49803), and jina-embeddings-v5-text-nano (#50688).
* **PyTorch 2.13.0 upgrade** along with torchvision 0.28.0 and Triton 3.7.1 (#48155) — this is a breaking environment change; XPU (#48677) and CPU (#50412) followed to torch 2.13 as well.
* **FlashAttention 4 integration deepens on SM100**: FP8 KV cache support (#42569) and headdim-256 support (#42669), backed by a new JIT warmup infrastructure (#47451) and runner-owned Triton kernel warmup (#49903) that remove first-request compilation stalls.
* **DeepSeek-V4 performance push**: sequence parallelism (#46789), ~2x kernel improvement by skipping empty c128 launches (#48957), 3.4% E2E TTFT from skipping unneeded topk/router (#49486), 3.9% E2E TTFT from workspace reuse (#49236), 1.88x kernel from removing a redundant full kernel (#50298), adaptive topk width (1.0% E2E, #50004), 448 MiB GPU memory saved in the PP buffer (#50312), a compact MXFP4 indexer KV cache (#48993), and removal of sparse-MLA q-head padding on FlashInfer >= 0.6.14 (#48047).
* **Model Runner V2 expands to non-generative workloads**: encoder-only attention (#49331), sequence pooling for embedding/classification (#48791), encoder token classification (#50293) and token embedding (#50574), BGE-M3 pooling (#50661), multimodal on CPU (#50073), a multi-layer MTP speculator (#48892), and PCP now selects MRV2 (#50034).
* **Resilient large-scale serving**: a (simplified) fault tolerance framework for DP+EP external load-balancer deployments (#44428) and async preparation for elastic EP scaling (#47288).
* **Disaggregation for hybrid models**: NIXL P/D for hybrid MLA+SSM models (#49762), heterogeneous P/D block sizes for hybrid models (#49612), and MoRIIO heterogeneous TP<->DP prefill/decode read routing (#46116).
* **Rust frontend grows a gRPC control plane**: engine-aware health reporting (#48992), abort control (#49255), server and model discovery (#49491), KV event source discovery (#50033), plus `vllm-bench` integrated into the `vllm` CLI (#48930).
* **Early next-gen hardware enablement**: `sm_107` target for NVIDIA Rubin (#49387) with NVLink all-reduce paths on SM107 (#49647), and ROCm gfx1250 architecture enabled (#46516).

###### Model Support
* Kimi K3: new model (#50000) with model files and kernels (#50089), Python frontend (#50093), Rust frontend (#50104), AttnRes kernels (#50090), DeepGEMM support (#50458), DSpark AR fusion (#50242), and optional shared-expert sharding (#50656).
* New models: Qwen3.5 text-only dense and MoE (#50210), K-EXAONE-2.0-750B-A37B (#50524), VaultGemma via Transformers backend (#49803), jina-embeddings-v5-text-nano with EuroBERT encoder backbone (#50688).
* Inkling: llm-compressor NVFP4 weights (#49258) and compressed-tensors dynamic FP8 (#48876).
* Multimodal: VidCom2 video token pruning (#47750), EVS for Qwen3.5 (#48912), ViT CUDA graph for Gemma-4 (#46837), Cosmos3 FP8 ModelOpt/Diffusers remapping (#48952), MiniMax-M3 MSA speculative decode verification (#50032) and default video processor (#50305), DeepSeek-OCR-2 TTFT optimization (#49531), longer max audio duration for MOSS-TD (#49403).
* Diffusion models: top_k and top_p sampling for DiffusionGemma (#45429).
* Transformers modeling backend: audio model support (#39330), improved `fx` tracer (#49957), fused residual-add + RMSNorm compilation

_[Truncated at 4000 characters — full notes: https://github.com/vllm-project/vllm/releases/tag/v0.27.0]_