Back to AI intel
重点
vLLM v0.30.0 Released with DeepSeek-V4.1-Flash Support
AI intel briefing
Core summary
One sentence to understand this update
vLLM version 0.30.0 has been released, featuring support for new models like DeepSeek-V4.1-Flash, which utilizes MXFP8 for KV storage.
Impact & opportunity
What this could mean
Builders can now leverage vLLM for high-performance inference with DeepSeek-V4.1-Flash, potentially improving efficiency with MXFP8 KV storage.
Source
View original