Back to AI intel
重点
vLLM v0.30.0 Adds DeepSeek-V4.1-Flash with MXFP8 KV Storage
AI intel briefing
Core summary
One sentence to understand this update
vLLM v0.30.0 is released, featuring 762 commits and new model support including DeepSeek-V4.1-Flash with MXFP8 KV storage, particularly optimized for SM100.
Impact & opportunity
What this could mean
Developers using vLLM can now deploy DeepSeek-V4.1-Flash with enhanced memory efficiency on compatible hardware, potentially reducing inference costs.
Source
View original