Speed test matrix

Tess-4-27B-NVFP4 on RTX 3090 24GB

vllm · 0.32.0 · NVFP44 contexts (0–56K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 56K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 29, 2026, 04:14 UTCEdited Oct 29, 2026, 12:41 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: NVFP4
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 migtissera/Tess-4-27B-NVFP4 --max-model-len 65536 --gpu-memory-utilization 0.94 --max-num-seqs 29 --enable-prefix-caching --trust-remote-code --language-model-only --kv-cache-dtype int4_per_token_head --attention-backend TRITON_ATTN --max-cudagraph-capture-size 16

Prefill

prompt tok/s by context length

8K context1,7844,590 ms TTFT

16K context1,48710,906 ms TTFT

32K context1,13528,460 ms TTFT

56K context836.067,464 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	41.996 ms TTFT	82.2109 ms TTFT	158.8242 ms TTFT	180.2Requests queued359 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
16K	37.6927 ms TTFT	70.71,405 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	34.22,524 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
56K	27.3Requests queued3,328 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	1,784.0	—	4,590 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,487.0	—	10,906 ms	—	—	1	—	—
Prefill	32K	1	Valid	1,135.0	—	28,460 ms	—	—	1	—	—
Prefill	56K	1	Valid	836.0	—	67,464 ms	—	—	1	—	—
Sustained decode	0	1	Valid	41.9	41.9	96 ms	23.8 ms	1.0	0	1,257	30.0 s
Sustained decode	0	2	Valid	82.2	41.2	109 ms	24.3 ms	2.0	0	2,466	30.0 s
Sustained decode	0	4	Valid	158.8	39.8	242 ms	25.1 ms	4.0	0	4,764	30.0 s
Sustained decode	0	8	Requests queued	180.2	33.6	359 ms	29.8 ms	6.3	0	5,404	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	37.6	37.7	927 ms	26.6 ms	1.0	0	1,128	30.0 s
Sustained decode	16K	2	Valid	70.7	35.0	1,405 ms	28.6 ms	2.0	0	2,122	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	34.2	34.2	2,524 ms	29.2 ms	1.0	0	1,026	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	—	—	—	—	—	—	—	—
Sustained decode	56K	1	Requests queued	27.3	30.1	3,328 ms	33.3 ms	0.9	0	818	30.0 s
Sustained decode	56K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	56K	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.