Speed test matrix

Fara1.5-4B-FP8 on RTX 3060 12GB

vllm · 0.32.0 · FP86 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, 03:16 UTC

Hardware

Type: Discrete GPU
GPU: RTX 3060
VRAM: 12 GB
CPU: Intel Xeon E5-2680 v4 (14C)
RAM: 32 GB
OS: Ubuntu 24.04.5

Engine

Engine: vllm
Version: 0.32.0
Quantization: FP8
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 3060 12GB, stock 170 W limit, PCIe Gen3 x16 | Intel Xeon E5-2680 v4 14C | 32 GB DDR4-2400 | Ubuntu 24.04.5, NVIDIA driver 580.178.04
vllm serve prithivMLmods/Fara1.5-4B-FP8 --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 context2,3753,430 ms TTFT

16K context1,9708,181 ms TTFT

32K context1,49421,483 ms TTFT

64K context1,00064,009 ms TTFT

128K context604.0211,894 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	49.780 ms TTFT	94.7116 ms TTFT	181.2231 ms TTFT	333.1394 ms TTFT	574.4634 ms TTFT	909.6988 ms TTFT	Over capacity	Over capacity
16K	44.81,261 ms TTFT	80.41,853 ms TTFT	136.93,028 ms TTFT	215.75,381 ms TTFT	301.110,151 ms TTFT	Over capacity	Over capacity	Over capacity
32K	40.81,536 ms TTFT	69.92,250 ms TTFT	109.53,658 ms TTFT	158.36,450 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
64K	34.51,295 ms TTFT	55.31,793 ms TTFT	78.32,817 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
128K	24.0Requests queued3,442 ms TTFT	31.5Requests queued4,950 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
248K	Over capacity	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	2,375.0	—	3,430 ms	—	—	1	—	—
Prefill	16K	1	Valid	1,970.0	—	8,181 ms	—	—	1	—	—
Prefill	32K	1	Valid	1,494.0	—	21,483 ms	—	—	1	—	—
Prefill	64K	1	Valid	1,000.0	—	64,009 ms	—	—	1	—	—
Prefill	128K	1	Valid	604.0	—	211,894 ms	—	—	1	—	—
Sustained decode	0	1	Valid	49.7	49.8	80 ms	20.1 ms	1.0	0	1,491	30.0 s
Sustained decode	0	2	Valid	94.7	47.5	116 ms	21.1 ms	2.0	0	2,842	30.0 s
Sustained decode	0	4	Valid	181.2	45.4	231 ms	22.0 ms	4.0	0	5,432	30.0 s
Sustained decode	0	8	Valid	333.1	41.8	394 ms	23.9 ms	8.0	0	9,992	30.0 s
Sustained decode	0	16	Valid	574.4	36.1	634 ms	27.7 ms	16.0	0	17,232	30.0 s
Sustained decode	0	32	Valid	909.6	28.6	988 ms	35.0 ms	32.0	0	27,296	30.0 s
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	44.8	44.8	1,261 ms	22.3 ms	1.0	0	1,342	30.0 s
Sustained decode	16K	2	Valid	80.4	39.6	1,853 ms	25.3 ms	2.0	0	2,411	30.0 s
Sustained decode	16K	4	Valid	136.9	32.7	3,028 ms	30.6 ms	4.0	0	4,104	30.0 s
Sustained decode	16K	8	Valid	215.7	24.2	5,381 ms	41.3 ms	8.0	0	6,464	30.0 s
Sustained decode	16K	16	Valid	301.1	15.1	10,151 ms	66.3 ms	16.0	0	9,024	30.0 s
Sustained decode	16K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	32K	1	Valid	40.8	40.8	1,536 ms	24.5 ms	1.0	0	1,222	30.0 s
Sustained decode	32K	2	Valid	69.9	34.3	2,250 ms	29.1 ms	2.0	0	2,094	30.0 s
Sustained decode	32K	4	Valid	109.5	25.9	3,658 ms	38.6 ms	4.0	0	3,284	30.0 s
Sustained decode	32K	8	Valid	158.3	17.5	6,450 ms	57.3 ms	8.0	0	4,744	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	34.5	34.6	1,295 ms	28.9 ms	1.0	0	1,036	30.0 s
Sustained decode	64K	2	Valid	55.3	27.4	1,793 ms	36.6 ms	2.0	0	1,658	30.0 s
Sustained decode	64K	4	Valid	78.3	18.9	2,817 ms	53.0 ms	4.0	0	2,348	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	Requests queued	24.0	26.5	3,442 ms	37.7 ms	0.9	0	719	30.0 s
Sustained decode	128K	2	Requests queued	31.5	18.5	4,950 ms	54.1 ms	1.8	0	946	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	Over capacity	—	—	—	—	0.7	0	0	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.