Ryan Zhou

mentions 2 type Person feed RSS

// recent coverage 2 mentions

12:21

2026-08-21

matthusby.github.io

artificial-intelligence

Real world(ish) DeepSeek V4 Flash performance on a single MI300X

A single AMD MI300X GPU can serve 32 concurrent coding agents running DeepSeek V4 Flash, delivering 582 generation tokens per second after tuning, according to a benchmark by developer Ryan Zhou. The …

13:10

2026-08-04

sourcefeed.dev

artificial-intelligence

DeepSeek V4 Flash on One AMD GPU Took Nine Patches

A single AMD MI300X GPU with 192 GB of HBM3 now serves DeepSeek's 284B-parameter DeepSeek-V4-Flash-0731 checkpoint in mixed FP4+FP8 format, requiring nine patch overlays against a vLLM ROCm nightly pl…

// co-occurs with top 8 entities

vLLM 2 AMD MI300X 1 DeepSeek V4 Flash 1 opencode 1 AMD Developer Cloud 1 DeepSeek 1 DeepSeek-V4-Flash-0731 1 AMD 1

// topics top 5 topics

artificial intelligence 2 large language models 2 ai infrastructure 2 ai tools 2 machine learning 1