Speed test matrix

Ornith-1.5-9B on RTX 3090 24GB

vllm · 0.32.0 · BF165 contexts (0–80K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 80K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 31, 2026, 00:34 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 ornith-ai/Ornith-1.5-9B --max-model-len 90112 --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 context6,9661,176 ms TTFT

16K context6,9182,346 ms TTFT

32K context6,4065,041 ms TTFT

64K context5,71911,265 ms TTFT

80K context5,40414,892 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	50.036 ms TTFT	97.644 ms TTFT	192.287 ms TTFT	378.8129 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
16K	48.4134 ms TTFT	92.6176 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	47.0117 ms TTFT	88.1162 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	44.5206 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
80K	43.4288 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	6,966.0	—	1,176 ms	—	—	1	—	—
Prefill	16K	1	Valid	6,918.0	—	2,346 ms	—	—	1	—	—
Prefill	32K	1	Valid	6,406.0	—	5,041 ms	—	—	1	—	—
Prefill	64K	1	Valid	5,719.0	—	11,265 ms	—	—	1	—	—
Prefill	80K	1	Valid	5,404.0	—	14,892 ms	—	—	1	—	—
Sustained decode	0	1	Valid	50.0	50.0	36 ms	20.0 ms	1.0	0	1,499	30.0 s
Sustained decode	0	2	Valid	97.6	48.8	44 ms	20.5 ms	2.0	0	2,928	30.0 s
Sustained decode	0	4	Valid	192.2	48.1	87 ms	20.8 ms	4.0	0	5,764	30.0 s
Sustained decode	0	8	Valid	378.8	47.5	129 ms	21.1 ms	8.0	0	11,360	30.0 s
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	48.4	48.4	134 ms	20.7 ms	1.0	0	1,451	30.0 s
Sustained decode	16K	2	Valid	92.6	46.3	176 ms	21.6 ms	2.0	0	2,776	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	47.0	47.0	117 ms	21.3 ms	1.0	0	1,410	30.0 s
Sustained decode	32K	2	Valid	88.1	44.1	162 ms	22.7 ms	2.0	0	2,642	30.0 s
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	—	—	—	—	—	—	—	—
Sustained decode	64K	1	Valid	44.5	44.6	206 ms	22.4 ms	1.0	0	1,335	30.0 s
Sustained decode	64K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	64K	4	Over capacity	—	—	—	—	—	—	—	—
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	80K	1	Valid	43.4	43.4	288 ms	23.0 ms	1.0	0	1,301	30.0 s
Sustained decode	80K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	128	Over capacity	—	—	—	—	—	—	—	—

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