DeepSeek-V2 MLA

part of DeepSeek V2

1 statements across 1 episodes · 0 bullish · 1 bearish · 1 people on the record · first statement Jan 19, 2025 by Yining Zhang · across every show →

Everything said about DeepSeek-V2 MLA, oldest first

Jan 19, 2025 negative

Assertion Supported

Zhang: vLLM does not support DeepSeek MLA while SGLang does

“Something like DeepSeq V-II, they proposed attention parent named MLA, multi-latent attention, and I think SGLAN is the only framework to support that. Maybe LightLM and TRTM also support, but VLM doesn't support.”

Yining Zhang Jan 19, 2025 ▶ 27:23 DeepSeek V3, SGLang, and the state of Open Model Inference in 2025 (Quantization, MoEs, Pricing)

Turn any episode into a week of clips.

This entire site, over 200 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

DeepSeek-V2 MLA

Everything said about DeepSeek-V2 MLA, oldest first

Turn any episode into a week of clips.