Speed test matrix

Qwen3.5-27B-AWQ-4bit on RTX 3090 24GB

vllm · 0.32.0 · AWQ 4-bit3 contexts (0–24K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 24K

by @Sandbergllm-inference-bench v0.8.0Measured Nov 1, 2026, 01:57 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: AWQ 4-bit
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. KV cache: fp8
# 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 cyankiwi/Qwen3.5-27B-AWQ-4bit --max-model-len 32768 --gpu-memory-utilization 0.94 --max-num-seqs 29 --enable-prefix-caching --trust-remote-code --language-model-only --kv-cache-dtype fp8 --max-cudagraph-capture-size 16

Prefill

prompt tok/s by context length

8K context1,8424,446 ms TTFT

16K context1,7789,122 ms TTFT

24K context1,70914,190 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	44.572 ms TTFT	83.0127 ms TTFT	160.2212 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
16K	43.1365 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
24K	42.6525 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	1,842.0	—	4,446 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,778.0	—	9,122 ms	—	—	1	—	—
Prefill	24K	1	Valid	1,709.0	—	14,190 ms	—	—	1	—	—
Sustained decode	0	1	Valid	44.5	44.5	72 ms	22.5 ms	1.0	0	1,335	30.0 s
Sustained decode	0	2	Valid	83.0	41.5	127 ms	24.1 ms	2.0	0	2,488	30.0 s
Sustained decode	0	4	Valid	160.2	40.1	212 ms	25.0 ms	4.0	0	4,804	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	43.1	43.1	365 ms	23.2 ms	1.0	0	1,293	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	42.6	42.6	525 ms	23.5 ms	1.0	0	1,277	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.