Speed test matrix

DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic on CMP 170HX 64GB

vllm · 0.32.0rc1.dev249+g18305948a · FP8 dynamic5 contexts (0–120K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 120K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 31, 2026, 15:32 UTC

Hardware

Type: Discrete GPU
GPU: CMP 170HX
VRAM: 64 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: FP8 dynamic
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. KV cache: fp8
# 1x 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 RedHatAI/DeepSeek-R1-Distill-Qwen-32B-FP8-dynamic --max-model-len 131072 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code  --kv-cache-dtype fp8

Prefill

prompt tok/s by context length

8K context1,8414,454 ms TTFT

16K context1,6489,849 ms TTFT

32K context1,39723,134 ms TTFT

64K context1,07260,209 ms TTFT

120K context762.0158,669 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	35.687 ms TTFT	70.2118 ms TTFT	138.3135 ms TTFT	268.1159 ms TTFT	509.1258 ms TTFT	Over capacity	Over capacity	Over capacity
16K	32.7175 ms TTFT	60.8246 ms TTFT	106.7340 ms TTFT	173.4435 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
32K	30.2258 ms TTFT	53.7347 ms TTFT	86.7514 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	26.2443 ms TTFT	43.6626 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
120K	21.6Requests queued693 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity

Requests queued: the server could not run all requests at once, so some waited in a queue. 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	1,841.0	—	4,454 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,648.0	—	9,849 ms	—	—	1	—	—
Prefill	32K	1	Valid	1,397.0	—	23,134 ms	—	—	1	—	—
Prefill	64K	1	Valid	1,072.0	—	60,209 ms	—	—	1	—	—
Prefill	120K	1	Valid	762.0	—	158,669 ms	—	—	1	—	—
Sustained decode	0	1	Valid	35.6	35.7	87 ms	28.0 ms	1.0	0	1,068	30.0 s
Sustained decode	0	2	Valid	70.2	35.2	118 ms	28.4 ms	2.0	0	2,106	30.0 s
Sustained decode	0	4	Valid	138.3	34.7	135 ms	28.8 ms	4.0	0	4,148	30.0 s
Sustained decode	0	8	Valid	268.1	33.7	159 ms	29.7 ms	8.0	0	8,040	30.0 s
Sustained decode	0	16	Valid	509.1	32.1	258 ms	31.2 ms	16.0	0	15,264	30.0 s
Sustained decode	0	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	32.7	32.8	175 ms	30.5 ms	1.0	0	982	30.0 s
Sustained decode	16K	2	Valid	60.8	30.4	246 ms	32.8 ms	2.0	0	1,824	30.0 s
Sustained decode	16K	4	Valid	106.7	26.8	340 ms	37.4 ms	4.0	0	3,202	30.0 s
Sustained decode	16K	8	Valid	173.4	21.7	435 ms	46.0 ms	8.0	0	5,200	30.0 s
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	30.2	30.2	258 ms	33.1 ms	1.0	0	904	30.0 s
Sustained decode	32K	2	Valid	53.7	26.9	347 ms	37.2 ms	2.0	0	1,610	30.0 s
Sustained decode	32K	4	Valid	86.7	21.7	514 ms	46.0 ms	4.0	0	2,600	30.0 s
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	26.2	26.3	443 ms	38.0 ms	1.0	0	787	30.0 s
Sustained decode	64K	2	Valid	43.6	21.9	626 ms	45.7 ms	2.0	0	1,308	30.0 s
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	120K	1	Requests queued	21.6	21.7	693 ms	46.1 ms	1.0	0	647	30.0 s
Sustained decode	120K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	120K	128	Over capacity	—	—	—	—	—	—	—	—

Requests queued: the server could not run all requests at once, so some waited in a queue. Over capacity: not run because this many requests at this context would exceed the server's KV cache.