Speed test matrix

granite-4.2-30b-nvfp4 on RTX 3090 24GB

vllm · 0.32.0 · NVFP44 contexts (0–48K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 48K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 29, 2026, 21:26 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 ibm-granite/granite-4.2-30b-nvfp4 --max-model-len 57344 --gpu-memory-utilization 0.94 --max-num-seqs 128 --enable-prefix-caching --trust-remote-code --kv-cache-dtype int4_per_token_head --attention-backend TRITON_ATTN

Prefill

prompt tok/s by context length

8K context1,3656,037 ms TTFT

16K context1,06515,323 ms TTFT

32K context746.043,503 ms TTFT

48K context574.084,779 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	40.375 ms TTFT	79.577 ms TTFT	152.761 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
16K	29.6142 ms TTFT	54.5222 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	23.3259 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
48K	19.3Requests queued426 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	1,365.0	—	6,037 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,065.0	—	15,323 ms	—	—	1	—	—
Prefill	32K	1	Valid	746.0	—	43,503 ms	—	—	1	—	—
Prefill	48K	1	Valid	574.0	—	84,779 ms	—	—	1	—	—
Sustained decode	0	1	Valid	40.3	40.5	75 ms	24.7 ms	1.0	0	1,210	30.0 s
Sustained decode	0	2	Valid	79.5	39.9	77 ms	25.1 ms	2.0	0	2,384	30.0 s
Sustained decode	0	4	Valid	152.7	38.5	61 ms	26.0 ms	4.0	0	4,580	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	29.6	29.6	142 ms	33.8 ms	1.0	0	887	30.0 s
Sustained decode	16K	2	Valid	54.5	27.2	222 ms	36.7 ms	2.0	0	1,634	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	23.3	23.3	259 ms	42.8 ms	1.0	0	699	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	48K	1	Requests queued	19.3	19.3	426 ms	51.8 ms	1.0	0	578	30.0 s
Sustained decode	48K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	48K	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.