Speed test matrix

Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-NVFP4-MTP on CMP 170HX 64GB

vllm · 0.32.0rc1.dev249+g18305948a · NVFP46 contexts (0–248K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 128K

by @Sandbergllm-inference-bench v0.8.0Measured Nov 1, 2026, 14:50 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: 56c8f43e1ef9423de0edbadd95a413ea80d62374
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 AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-Multimodal-NVFP4-MTP --max-model-len 262144 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code --language-model-only

Prefill

prompt tok/s by context length

8K context1,9444,190 ms TTFT

16K context1,8328,795 ms TTFT

32K context1,72018,655 ms TTFT

64K context1,52441,997 ms TTFT

128K context1,240103,152 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	39.1162 ms TTFT	76.3198 ms TTFT	147.2322 ms TTFT	278.5508 ms TTFT	490.0859 ms TTFT	791.51,439 ms TTFT	Over capacity	Over capacity
16K	37.6384 ms TTFT	71.0575 ms TTFT	130.7890 ms TTFT	227.21,548 ms TTFT	354.22,861 ms TTFT	Over capacity	Over capacity	Over capacity
32K	36.3704 ms TTFT	66.6965 ms TTFT	117.51,531 ms TTFT	192.02,659 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
64K	33.9843 ms TTFT	59.21,097 ms TTFT	97.91,672 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
128K	30.0990 ms TTFT	48.41,227 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
248K	23.1Requests queued2,622 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	1,944.0	—	4,190 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,832.0	—	8,795 ms	—	—	1	—	—
Prefill	32K	1	Valid	1,720.0	—	18,655 ms	—	—	1	—	—
Prefill	64K	1	Valid	1,524.0	—	41,997 ms	—	—	1	—	—
Prefill	128K	1	Valid	1,240.0	—	103,152 ms	—	—	1	—	—
Sustained decode	0	1	Valid	39.1	39.2	162 ms	25.5 ms	1.0	0	1,174	30.0 s
Sustained decode	0	2	Valid	76.3	38.2	198 ms	26.2 ms	2.0	0	2,288	30.0 s
Sustained decode	0	4	Valid	147.2	36.8	322 ms	27.1 ms	4.0	0	4,412	30.0 s
Sustained decode	0	8	Valid	278.5	34.9	508 ms	28.7 ms	8.0	0	8,352	30.0 s
Sustained decode	0	16	Valid	490.0	30.7	859 ms	32.6 ms	16.0	0	14,704	30.0 s
Sustained decode	0	32	Valid	791.5	24.8	1,439 ms	40.4 ms	32.0	0	23,744	30.0 s
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	37.6	37.6	384 ms	26.6 ms	1.0	0	1,127	30.0 s
Sustained decode	16K	2	Valid	71.0	35.4	575 ms	28.2 ms	2.0	0	2,130	30.0 s
Sustained decode	16K	4	Valid	130.7	32.3	890 ms	30.9 ms	4.0	0	3,920	30.0 s
Sustained decode	16K	8	Valid	227.2	27.7	1,548 ms	36.1 ms	8.0	0	6,816	30.0 s
Sustained decode	16K	16	Valid	354.2	20.9	2,861 ms	47.9 ms	16.0	0	10,624	30.0 s
Sustained decode	16K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	1	Valid	36.3	36.3	704 ms	27.5 ms	1.0	0	1,088	30.0 s
Sustained decode	32K	2	Valid	66.6	33.1	965 ms	30.2 ms	2.0	0	1,996	30.0 s
Sustained decode	32K	4	Valid	117.5	28.8	1,531 ms	34.7 ms	4.0	0	3,524	30.0 s
Sustained decode	32K	8	Valid	192.0	22.9	2,659 ms	43.7 ms	8.0	0	5,752	30.0 s
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	33.9	34.0	843 ms	29.4 ms	1.0	0	1,017	30.0 s
Sustained decode	64K	2	Valid	59.2	29.5	1,097 ms	33.9 ms	2.0	0	1,774	30.0 s
Sustained decode	64K	4	Valid	97.9	24.0	1,672 ms	41.6 ms	4.0	0	2,936	30.0 s
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	128K	1	Valid	30.0	30.2	990 ms	33.2 ms	1.0	0	900	30.0 s
Sustained decode	128K	2	Valid	48.4	24.3	1,227 ms	41.2 ms	2.0	0	1,450	30.0 s
Sustained decode	128K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	1	Requests queued	23.1	24.9	2,622 ms	40.2 ms	0.9	0	694	30.0 s
Sustained decode	248K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	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.