# vLLM v0.30.0 — engine restarts skip the disk with a GPU weight cache

> Fast Start keeps post-quantized weights resident in GPU memory, so a restarting vLLM engine maps them instead of reloading them.

- Org: vLLM
- Date: 2026-09-22
- Category: repo, tool
- Source: https://github.com/vllm-project/vllm/releases/tag/v0.30.0
- Canonical page: https://ai-tldr.dev/releases/vllm-v0-30-0/

## In short

vLLM v0.30.0 adds Fast Start, a per-GPU daemon that holds quantized weights in GPU memory so a restarting engine maps them over CUDA IPC instead of reloading from disk. The release lands 762 commits from 315 contributors.

## What it is

Fast Start is the headline addition in vLLM v0.30.0: a persistent per-GPU daemon that holds post-quantized, tensor-parallel-sharded weights in GPU memory. Engines started with `--load-format ipc_cache` map those weights over CUDA IPC rather than reading a checkpoint off disk. The cache covers FP4 checkpoints and multi-node tensor parallelism. A second new feature, Gumbel-max watermarking, can mark and later detect vLLM-generated text.

## How it works

Startup time in v0.30.0 also drops because garbage collection is frozen during CUDA graph capture: capture falls from 12s to 2s and engine init from 28.9s to 8.2s on an H200. A new tier called HiSparse spills KV pages to pinned host memory when a sparse-MLA model runs out of GPU room, then serves top-k misses from a per-request GPU hot buffer. Kimi K3 decoding gains 5.2–7.7% end-to-end throughput after a gather/scatter step was removed from its KDA path.

## Why it matters

Restart cost is what makes autoscaling a serving fleet expensive, and mapping cached weights removes most of it — useful for anyone who redeploys often or runs reinforcement-learning loops that tear engines down between steps. The upgrade is not free, though: scale-out endpoints are now opt-in, GPTQ activation ordering is gone, and YaRN now matches Transformers, so vendor YaRN aliases no longer re-scale `max_model_len`.

## Who it's for

inference and platform engineers

## Try it

pip install -U vllm && vllm serve <model> --load-format ipc_cache

## Quick facts

- Maker: vLLM project
- Version: v0.30.0
- Released: September 22, 2026
- License: Apache-2.0
- Contributors: 315 across 762 commits
- Headline feature: Fast Start weight-cache daemon
- Breaking change: Scale-out endpoints need --enable-scale-out

## FAQ

### What breaks when upgrading to vLLM v0.30.0?

vLLM v0.30.0 makes scale-out endpoints opt-in on plain `vllm serve` behind `--enable-scale-out`, replacing the old environment variable. GPTQ activation ordering (`g_idx`) is removed, the `all` Mamba cache mode is deprecated, and `python -m vllm.entrypoints.grpc_server` gives way to `vllm serve --grpc`. YaRN now follows Transformers, so vendor aliases stop re-scaling context length.

### Which new models does vLLM v0.30.0 serve?

New in vLLM v0.30.0 are DeepSeek-V4.1-Flash, which stores its whole KV cache in MXFP8, plus DeepSeek-V4-Flash-Vision-Exp with ROCm and LoRA support, GLM-5.3-Flash, K2-Horizon with its reasoning and tool parsers, Cohere Compass, Bailing V3 VL, and Nanbeige4.2 through the Transformers backend. A DeepSeek-V4 CPU backend with AVX512 and AMX kernels also lands.

### How does watermarking in vLLM v0.30.0 work for users?

Watermarking in vLLM v0.30.0 uses Gumbel-max sampling with a keyed pseudorandom function, and the release ships an example endpoint that detects the mark later. Callers can opt out per request, and a dual-key variant keeps the watermark compatible with speculative decoding. The Rust frontend forwards the per-request controls.

### Can vLLM v0.30.0 run without NVLink between GPUs?

Yes — vLLM v0.30.0 adds an opt-in FlashInfer PCIe IPC all-reduce aimed at boxes without NVLink. The same large-scale serving work brings prefill and decode context parallelism to sparse-MLA models, Elastic EP that reuses CUDA graphs across reconfiguration, and encoder-cache sharing over NIXL and Mooncake.

## Links

- [Release Notes](https://github.com/vllm-project/vllm/releases/tag/v0.30.0)
- [Repository](https://github.com/vllm-project/vllm)
- [Documentation](https://docs.vllm.ai/en/latest/)
- [PyPI](https://pypi.org/project/vllm/)