minimax-m3.1POST
MiniMax M3.1 Flash on 4x DGX Spark: 84 tok/s decode with vLLM
A user ran MiniMax M3.1 Flash locally on four DGX Sparks with vLLM and DSpark speculative decoding, handling six ~100K-token prompts at 84 tok/s combined decode and ~1,480 tok/s prefill.
dextune
@dextuneArchived 3 days agoOriginally posted 2026.10.24
dextune @dextuneOriginally posted 2026.10.24
MiniMax M3.1 Flash on 4x DGX Spark @ToNYD2WiLD reports running it across four DGX Sparks with vLLM and DSpark speculative decoding. In the 2026-10-23 run, the setup handled six ~100K-token prompts at once, with 84 tok/s combined decode and about 1,480 tok/s prefill. https://github.com/tonyd2wild/MiniMax-M3.1-Flash-TP4-DGX-Sparks
x.com/dextune/status/2113841813614473354Archived 2026.10.25 12:07