Speed test matrix

MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10 on CMP 170HX 64GB ×2

vllm · 0.32.0rc1.dev249+g18305948a · NVFP45 contexts (0–88K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 88K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 29, 2026, 11:23 UTC

Hardware

Type: Discrete GPU
GPU: CMP 170HX
GPU count: 2
VRAM: 64 GB (2 × 32 GB)
CPU: 2x Intel Xeon E5-2630 v3
RAM: 64 GB
OS: Ubuntu 26.04.1

Engine

Engine: vllm
Version: 0.32.0rc1.dev249+g18305948a
Quantization: NVFP4
Backend: cuda
Repository: github.com/vllm-project/vllm
Build: ghcr.io/morrowmake/vllm-cmp170hx@sha256:5b45c4d6feafefe2eba962375b0a9a0b64fd22b9e05abb4d1966f14a32a2db79

Run

Served model: bench
Revision: main
Benchmark tool: llm-inference-bench v0.8.0
Decode test: Fixed duration
Prefill mode: Measured during decode warm-up
Time per measurement: 30 s
Max output: 8,192 tokens

Launch command

# Untuned out-of-the-box baseline. Spec decode: none. Layout: PP2 (stock vLLM flag)
# 2x NVIDIA CMP 170HX 64GB (cmpunlocker), 180 W limit, PCIe Gen2 x4 | 2x Intel Xeon E5-2630 v3 | 64 GB DDR4 | Ubuntu 26.04.1, NVIDIA driver 610.57.04 | vLLM: Morrowmake vllm-cmp170hx image
vllm serve scottgl/MiniMax-M2.7-REAP-172B-A10B-NVFP4-GB10 --max-model-len 98304 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code --pipeline-parallel-size 2

Prefill

prompt tok/s by context length

8K context4,0251,985 ms TTFT

16K context3,9883,959 ms TTFT

32K context3,3679,329 ms TTFT

64K context2,49025,164 ms TTFT

88K context2,06841,633 ms TTFT

Sustained decode

total tok/s · rows are context length, columns are simultaneous requests

Context	1	2	4	8	16	32	64	128
0	89.0121 ms TTFT	126.2130 ms TTFT	226.1106 ms TTFT	274.9135 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
16K	71.2213 ms TTFT	124.0231 ms TTFT	168.9348 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	59.5282 ms TTFT	77.3386 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	44.4525 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
88K	37.1739 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity

Over capacity: not run because this many requests at this context would exceed the server's KV cache.

TTFT is measured after the benchmark warms the server's prompt cache, so it mostly excludes prefill.

All measurements

Latencies are medians. Total tok/s is output across all simultaneous requests.

Phase	Context	Requests at once	Status	Total tok/s	Per request tok/s	TTFT	Time between tokens	Average active requests	Completed requests	Output tokens	Measured for
Prefill	8K	1	Valid	4,025.0	—	1,985 ms	—	—	1	—	—
Prefill	16K	1	Valid	3,988.0	—	3,959 ms	—	—	1	—	—
Prefill	32K	1	Valid	3,367.0	—	9,329 ms	—	—	1	—	—
Prefill	64K	1	Valid	2,490.0	—	25,164 ms	—	—	1	—	—
Prefill	88K	1	Valid	2,068.0	—	41,633 ms	—	—	1	—	—
Sustained decode	0	1	Valid	89.0	89.5	121 ms	11.2 ms	1.0	0	2,669	30.0 s
Sustained decode	0	2	Valid	126.2	63.5	130 ms	15.7 ms	2.0	0	3,786	30.0 s
Sustained decode	0	4	Valid	226.1	56.8	106 ms	17.6 ms	4.0	0	6,781	30.0 s
Sustained decode	0	8	Valid	274.9	34.9	135 ms	28.7 ms	8.0	0	8,248	30.0 s
Sustained decode	0	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	71.2	71.4	213 ms	14.0 ms	1.0	0	2,136	30.0 s
Sustained decode	16K	2	Valid	124.0	62.2	231 ms	16.1 ms	2.0	0	3,720	30.0 s
Sustained decode	16K	4	Valid	168.9	42.3	348 ms	23.6 ms	4.0	0	5,066	30.0 s
Sustained decode	16K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	1	Valid	59.5	59.6	282 ms	16.8 ms	1.0	0	1,783	30.0 s
Sustained decode	32K	2	Valid	77.3	38.8	386 ms	25.8 ms	2.0	0	2,318	30.0 s
Sustained decode	32K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	1	Valid	44.4	44.5	525 ms	22.5 ms	1.0	0	1,331	30.0 s
Sustained decode	64K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	1	Valid	37.1	37.2	739 ms	26.9 ms	1.0	0	1,112	30.0 s
Sustained decode	88K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	88K	128	Over capacity	—	—	—	—	—	—	—	—

Over capacity: not run because this many requests at this context would exceed the server's KV cache.