deepseek-v4.1IMAGE
DeepSeek V4.1 Flash on 4x DGX Spark: 165.7 tok/s, 98.6 on 69-scenario tool eval
The author deployed DeepSeek V4.1 Flash 1017 with vLLM on 4x DGX Spark, reporting 165.7 tok/s throughput and 98.6 on a self-built 69-scenario tool-use eval, with the recipe published.
Tech2Wild
@Tech2WildArchived 11 days agoOriginally posted 2026.10.18
Tech2Wild @Tech2WildOriginally posted 2026.10.18
I ran my own benchmark on DeepSeek V4.1 Flash 1017 and here's what it scored on 4x DGX Spark 🔥 ⚡ 165.7 tok/s throughput 🎯 98.6 on my 69-scenario tool eval Full recipe 👇 https://github.com/tonyd2wild/DeepSeek-V4.1-Flash-vLLM-DGX-Spark
x.com/tech2wild/status/2111506332004086245Archived 2026.10.18 16:30