Speed test matrix

Ornith-1.5-9B on RTX 3090 24GB

vllm · 0.32.0 · BF166 contexts (0–248K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 128K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 31, 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: none. KV cache: int4_per_token_head
# 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 262144 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code --language-model-only --kv-cache-dtype int4_per_token_head --attention-backend TRITON_ATTN

Prefill

prompt tok/s by context length

8K context6,3981,281 ms TTFT

16K context5,4712,966 ms TTFT

32K context4,3087,496 ms TTFT

64K context3,02421,307 ms TTFT

128K context1,88668,232 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	49.242 ms TTFT	96.346 ms TTFT	189.591 ms TTFT	372.7137 ms TTFT	693.5258 ms TTFT	1,186.0359 ms TTFT	Over capacity	Over capacity
16K	46.1485 ms TTFT	89.4711 ms TTFT	165.11,177 ms TTFT	304.22,099 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
32K	43.5609 ms TTFT	83.6888 ms TTFT	145.51,448 ms TTFT	256.02,569 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
64K	39.0678 ms TTFT	73.9969 ms TTFT	119.21,499 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
128K	32.41,919 ms TTFT	60.12,760 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
248K	22.5Requests queued3,031 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity

Requests queued: the server could not run all requests at once, so some waited in a queue. 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	6,398.0	—	1,281 ms	—	—	1	—	—
Prefill	16K	1	Valid	5,471.0	—	2,966 ms	—	—	1	—	—
Prefill	32K	1	Valid	4,308.0	—	7,496 ms	—	—	1	—	—
Prefill	64K	1	Valid	3,024.0	—	21,307 ms	—	—	1	—	—
Prefill	128K	1	Valid	1,886.0	—	68,232 ms	—	—	1	—	—
Sustained decode	0	1	Valid	49.2	49.2	42 ms	20.3 ms	1.0	0	1,475	30.0 s
Sustained decode	0	2	Valid	96.3	48.1	46 ms	20.8 ms	2.0	0	2,886	30.0 s
Sustained decode	0	4	Valid	189.5	47.4	91 ms	21.1 ms	4.0	0	5,684	30.0 s
Sustained decode	0	8	Valid	372.7	46.7	137 ms	21.4 ms	8.0	0	11,176	30.0 s
Sustained decode	0	16	Valid	693.5	43.5	258 ms	23.0 ms	16.0	0	20,800	30.0 s
Sustained decode	0	32	Valid	1,186.0	37.2	359 ms	26.9 ms	32.0	0	35,584	30.0 s
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	46.1	46.2	485 ms	21.7 ms	1.0	0	1,384	30.0 s
Sustained decode	16K	2	Valid	89.4	44.4	711 ms	22.5 ms	2.0	0	2,680	30.0 s
Sustained decode	16K	4	Valid	165.1	40.6	1,177 ms	24.7 ms	4.0	0	4,952	30.0 s
Sustained decode	16K	8	Valid	304.2	36.5	2,099 ms	27.4 ms	8.0	0	9,120	30.0 s
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	43.5	43.5	609 ms	23.0 ms	1.0	0	1,304	30.0 s
Sustained decode	32K	2	Valid	83.6	41.5	888 ms	24.1 ms	2.0	0	2,506	30.0 s
Sustained decode	32K	4	Valid	145.5	35.6	1,448 ms	28.1 ms	4.0	0	4,363	30.0 s
Sustained decode	32K	8	Valid	256.0	30.4	2,569 ms	32.9 ms	8.0	0	7,680	30.0 s
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	39.0	39.1	678 ms	25.6 ms	1.0	0	1,170	30.0 s
Sustained decode	64K	2	Valid	73.9	36.7	969 ms	27.2 ms	2.0	0	2,216	30.0 s
Sustained decode	64K	4	Valid	119.2	29.2	1,499 ms	34.2 ms	4.0	0	3,576	30.0 s
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	128K	1	Valid	32.4	32.4	1,919 ms	30.9 ms	1.0	0	971	30.0 s
Sustained decode	128K	2	Valid	60.1	29.4	2,760 ms	34.1 ms	2.0	0	1,800	30.0 s
Sustained decode	128K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	128K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	1	Requests queued	22.5	24.5	3,031 ms	40.8 ms	0.9	0	674	30.0 s
Sustained decode	248K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	248K	128	Over capacity	—	—	—	—	—	—	—	—

Requests queued: the server could not run all requests at once, so some waited in a queue. Over capacity: not run because this many requests at this context would exceed the server's KV cache.