Speed test matrix

Qwen3-0.6B on RTX 3090 24GB

vllm · 0.32.0 · BF163 contexts (0–32K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 31, 2026, 05:40 UTC

Hardware

Type: Discrete GPU
GPU: RTX 3090
VRAM: 24 GB
CPU: Intel Core i7-9700K (8C)
RAM: 32 GB
OS: Ubuntu 24.04.5

Engine

Engine: vllm
Version: 0.32.0
Quantization: BF16
Backend: cuda
Repository: github.com/vllm-project/vllm
Build: vllm/vllm-openai:v0.32.0-x86_64

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 RTX 3090 24GB (Zagitta/gpu-unlock), 420 W limit, PCIe Gen3 x16 | Intel Core i7-9700K 8C | 32 GB DDR4 | Ubuntu 24.04.5, NVIDIA driver 580.178.04
vllm serve Qwen/Qwen3-0.6B --max-model-len 40960 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code

Prefill

prompt tok/s by context length

8K context39.4K209 ms TTFT

16K context29.8K549 ms TTFT

32K context18.5K1,757 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	321.913 ms TTFT	518.823 ms TTFT	779.325 ms TTFT	1,180.832 ms TTFT	1,756.740 ms TTFT	Over capacity	Over capacity	Over capacity
16K	192.756 ms TTFT	271.074 ms TTFT	435.8120 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	139.8104 ms TTFT	181.3170 ms TTFT	277.7196 ms TTFT	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	39.4K	—	209 ms	—	—	1	—	—
Prefill	16K	1	Valid	29.8K	—	549 ms	—	—	1	—	—
Prefill	32K	1	Valid	18.5K	—	1,757 ms	—	—	1	—	—
Sustained decode	0	1	Valid	321.9	338.8	13 ms	3.0 ms	1.0	1	9,657	30.0 s
Sustained decode	0	2	Valid	518.8	301.2	23 ms	3.4 ms	2.0	2	15,564	30.0 s
Sustained decode	0	4	Valid	779.3	213.5	25 ms	4.7 ms	4.0	0	23,376	30.0 s
Sustained decode	0	8	Valid	1,180.8	166.8	32 ms	6.0 ms	8.0	0	35,416	30.0 s
Sustained decode	0	16	Valid	1,756.7	126.8	40 ms	7.9 ms	16.0	0	52,704	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	192.7	194.3	56 ms	5.1 ms	1.0	0	5,780	30.0 s
Sustained decode	16K	2	Valid	271.0	137.2	74 ms	7.3 ms	2.0	0	8,129	30.0 s
Sustained decode	16K	4	Valid	435.8	110.3	120 ms	9.1 ms	4.0	0	13,072	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	139.8	140.5	104 ms	7.1 ms	1.0	0	4,193	30.0 s
Sustained decode	32K	2	Valid	181.3	91.0	170 ms	11.0 ms	2.0	0	5,438	30.0 s
Sustained decode	32K	4	Valid	277.7	69.6	196 ms	14.4 ms	4.0	0	8,331	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	—	—	—	—	—	—	—	—

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