llamaperf

Minimax M2.1 230B (10B active)

on 2× RTX 4090D 48GB · llama.cpp · 65,536 ctx

Tone: positive
Oct 9, 2026
Throughput
20.1 t/s gen · 67.1 tokens/s prompt processing (prefill)Prompt processing is how fast the input is read, not how long you wait for the first word. How to measure time to first token.
Quant
UD-Q4_K_XL (GGUF)
System RAM
256 GB
VRAM reported
72 GB

Summary

User reports Minimax M2.1 at 20.05 t/s generation and 67.11 t/s prompt eval on an RTX 4090D 48GB plus an RTX 3090. Setup is llama.cpp build 7542 with UD-Q4_K_XL GGUF at 65536 context, running in router mode on a Xeon W5-3425 with 256GB DDR5. User notes speed is lower at larger context and says it works great.

Compare this run

Medians of plain runs: one device, one request, no speculative decoding, the whole model in the device's memory. The report above is one setup, so read it beside them and not as one of them.

Minimax M2.1 230B · 10B active at 4-bit by hardware

Bars are medians on one scale, 0 to 25.0 tokens/s. Press ▶ to watch text arrive at a row's speed.

Every Minimax M2.1 report →