Speed test matrix

Llama-3.3-70B-Instruct-NVFP4 on CMP 170HX 64GB

vllm · 0.32.0rc1.dev249+g18305948a · NVFP45 contexts (0–96K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 96K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 29, 2026, 21:10 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: 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
# 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 nvidia/Llama-3.3-70B-Instruct-NVFP4 --max-model-len 106496 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code

Prefill

prompt tok/s by context length

8K context702.011,691 ms TTFT

16K context637.025,482 ms TTFT

32K context563.057,284 ms TTFT

64K context460.0139,972 ms TTFT

96K context389.0248,130 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	21.5181 ms TTFT	43.0211 ms TTFT	85.2174 ms TTFT	165.0213 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
16K	20.3181 ms TTFT	38.4272 ms TTFT	69.3412 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	19.0Requests queued248 ms TTFT	34.8Requests queued364 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	17.1Requests queued464 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
96K	15.6Requests queued653 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	702.0	—	11,691 ms	—	—	1	—	—
Prefill	16K	1	Valid	637.0	—	25,482 ms	—	—	1	—	—
Prefill	32K	1	Valid	563.0	—	57,284 ms	—	—	1	—	—
Prefill	64K	1	Valid	460.0	—	139,972 ms	—	—	1	—	—
Prefill	96K	1	Valid	389.0	—	248,130 ms	—	—	1	—	—
Sustained decode	0	1	Valid	21.5	21.5	181 ms	46.5 ms	1.0	0	644	30.0 s
Sustained decode	0	2	Valid	43.0	21.5	211 ms	46.5 ms	2.0	0	1,288	30.0 s
Sustained decode	0	4	Valid	85.2	21.3	174 ms	46.9 ms	4.0	0	2,556	30.0 s
Sustained decode	0	8	Valid	165.0	20.7	213 ms	48.3 ms	8.0	0	4,944	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	20.3	20.4	181 ms	49.1 ms	1.0	0	610	30.0 s
Sustained decode	16K	2	Valid	38.4	19.3	272 ms	51.9 ms	2.0	0	1,153	30.0 s
Sustained decode	16K	4	Valid	69.3	17.4	412 ms	57.6 ms	4.0	0	2,080	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	Requests queued	19.0	19.0	248 ms	52.6 ms	1.0	0	570	30.0 s
Sustained decode	32K	2	Requests queued	34.8	17.4	364 ms	57.4 ms	2.0	0	1,042	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	Requests queued	17.1	17.1	464 ms	58.5 ms	1.0	0	511	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	96K	1	Requests queued	15.6	15.7	653 ms	63.9 ms	1.0	0	468	30.0 s
Sustained decode	96K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	96K	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.