Speed test matrix

Tess-4-9B 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 30, 2026, 11:11 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: MTP, 2 tokens lookahead
# 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 migtissera/Tess-4-9B --max-model-len 40960 --gpu-memory-utilization 0.94 --max-num-seqs 102 --enable-prefix-caching --trust-remote-code --language-model-only --speculative-config {"method":"mtp","num_speculative_tokens":2}

Prefill

prompt tok/s by context length

8K context6,9191,183 ms TTFT

16K context6,6292,447 ms TTFT

32K context6,1015,292 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	63.738 ms TTFT	131.276 ms TTFT	258.496 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
16K	52.5226 ms TTFT	108.2305 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	46.7228 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	6,919.0	—	1,183 ms	—	—	1	—	—
Prefill	16K	1	Valid	6,629.0	—	2,447 ms	—	—	1	—	—
Prefill	32K	1	Valid	6,101.0	—	5,292 ms	—	—	1	—	—
Sustained decode	0	1	Valid	63.7	65.1	38 ms	15.4 ms	1.0	0	1,910	30.0 s
Sustained decode	0	2	Valid	131.2	66.0	76 ms	15.2 ms	2.0	0	3,935	30.0 s
Sustained decode	0	4	Valid	258.4	64.7	96 ms	15.5 ms	4.0	0	7,749	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	52.5	53.2	226 ms	18.8 ms	1.0	0	1,575	30.0 s
Sustained decode	16K	2	Valid	108.2	54.1	305 ms	18.5 ms	2.0	0	3,243	30.0 s
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	32K	1	Valid	46.7	46.9	228 ms	21.3 ms	1.0	0	1,401	30.0 s
Sustained decode	32K	2	Over capacity	—	—	—	—	—	—	—	—
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	—	—	—	—	—	—	—	—

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