Speed test matrix

Qwen3.8-27B-INT4 on RTX 3090 24GB

vllm · 0.32.0 · INT4 W4A16 g1283 contexts (0–24K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 24K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 29, 2026, 14:36 UTC

Hardware

Type: Discrete GPU
GPU: RTX 3090
VRAM: 24 GB
CPU: AMD Ryzen 9 7900X (12C)
RAM: 32 GB
OS: Ubuntu 24.04.5

Engine

Engine: vllm
Version: 0.32.0
Quantization: INT4 W4A16 g128
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), 350 W limit (VBIOS max), PCIe Gen4 x16 | AMD Ryzen 9 7900X 12C | 32 GB DDR5 | Ubuntu 24.04.5, NVIDIA driver 595.99.02
vllm serve RedHatAI/Qwen3.8-27B-INT4 --max-model-len 32768 --gpu-memory-utilization 0.94 --max-num-seqs 54 --enable-prefix-caching --trust-remote-code --language-model-only --max-cudagraph-capture-size 16

Prefill

prompt tok/s by context length

8K context1,8454,463 ms TTFT

16K context1,8029,028 ms TTFT

24K context1,76313,778 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	46.987 ms TTFT	86.8142 ms TTFT	167.0415 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
16K	44.3379 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
24K	43.1524 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	1,845.0	—	4,463 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,802.0	—	9,028 ms	—	—	1	—	—
Prefill	24K	1	Valid	1,763.0	—	13,778 ms	—	—	1	—	—
Sustained decode	0	1	Valid	46.9	46.9	87 ms	21.3 ms	1.0	0	1,408	30.0 s
Sustained decode	0	2	Valid	86.8	43.5	142 ms	23.0 ms	2.0	0	2,604	30.0 s
Sustained decode	0	4	Valid	167.0	41.8	415 ms	23.9 ms	4.0	0	5,008	30.0 s
Sustained decode	0	8	Over capacity	—	—	—	—	—	—	—	—
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	44.3	44.3	379 ms	22.6 ms	1.0	0	1,328	30.0 s
Sustained decode	16K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	4	Over capacity	—	—	—	—	—	—	—	—
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	24K	1	Valid	43.1	43.1	524 ms	23.2 ms	1.0	0	1,293	30.0 s
Sustained decode	24K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	24K	128	Over capacity	—	—	—	—	—	—	—	—

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