# vLLM 0.31.0 improves DeepSeek-V4.1-Flash performance and large-scale serving

> **Key takeaway:** vLLM 0.31.0 introduces significant optimizations for DeepSeek models, enhanced speculative decoding, and new large-scale serving backends.

**Published:** 2026-10-06T06:49:35+00:00 
**Source:** vLLM High-Throughput Serving Engine 
**Category:** ai-local-edge 
**Canonical URL:** https://fosswire.org/news/vllm-0310-improves-deepseek-v41-flash-performance-and-large-.html 

## Summary
The vLLM 0.31.0 release includes 717 commits and introduces deep optimizations for DeepSeek-V4.1-Flash, including FlashMLA mega attention with NVFP4 compressed KV cache. The update also features Model Runner V2 with improved speculative decoding, new large-scale serving backends like MoonEP, and a new preload CLI to keep weights resident in GPU memory across restarts.

## Technical highlights of vLLM 0.31.0
- **DeepSeek-V4.1-Flash optimizations**: Implementation of FlashMLA mega attention, DeepGEMM sparse MQA logits, and Mega-Gate fusing.
- **Model Runner V2**: Supports draft-model speculative decoding and custom logits processors.
- **Large-scale serving**: Adds MoonEP balanced EP all2all backend and prefill context parallelism with data parallelism.
- **Fast restart capability**: A new `vllm preload` CLI uses a weight-cache daemon to maintain post-quantized weights in GPU memory.
- **Security enhancements**: Implements request-based gating for multimodal kwargs and prevents LoRA name/cache_salt collisions.

## Why it matters
This release significantly advances the efficiency of serving massive Mixture-of-Experts (MoE) models like DeepSeek on high-end hardware.