Speed test matrix

Fara1.5-4B-AutoRound-W4A16-RTN on RTX 3060 12GB

vllm · 0.32.0 · AutoRound W4A166 contexts (0–176K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 128K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 30, 2026, 21:05 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: AutoRound W4A16
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 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 LeaderboardModel1/Fara1.5-4B-AutoRound-W4A16-RTN --max-model-len 188416 --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 context2,7902,920 ms TTFT

16K context2,5806,247 ms TTFT

32K context2,29813,965 ms TTFT

64K context1,86934,265 ms TTFT

128K context1,35494,489 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	85.2104 ms TTFT	163.3168 ms TTFT	301.0212 ms TTFT	523.9340 ms TTFT	820.4576 ms TTFT	Over capacity	Over capacity	Over capacity
16K	74.2253 ms TTFT	127.9358 ms TTFT	199.9533 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	65.7432 ms TTFT	104.8593 ms TTFT	148.8893 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	53.4489 ms TTFT	76.4673 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
128K	38.8852 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
176K	32.0992 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	2,790.0	—	2,920 ms	—	—	1	—	—
Prefill	16K	1	Valid	2,580.0	—	6,247 ms	—	—	1	—	—
Prefill	32K	1	Valid	2,298.0	—	13,965 ms	—	—	1	—	—
Prefill	64K	1	Valid	1,869.0	—	34,265 ms	—	—	1	—	—
Prefill	128K	1	Valid	1,354.0	—	94,489 ms	—	—	1	—	—
Sustained decode	0	1	Valid	85.2	85.5	104 ms	11.7 ms	1.0	0	2,556	30.0 s
Sustained decode	0	2	Valid	163.3	82.1	168 ms	12.2 ms	2.0	0	4,898	30.0 s
Sustained decode	0	4	Valid	301.0	75.8	212 ms	13.2 ms	4.0	0	9,028	30.0 s
Sustained decode	0	8	Valid	523.9	66.2	340 ms	15.1 ms	8.0	0	15,718	30.0 s
Sustained decode	0	16	Valid	820.4	52.1	576 ms	19.2 ms	16.0	0	24,608	30.0 s
Sustained decode	0	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	0	128	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	16K	1	Valid	74.2	74.4	253 ms	13.4 ms	1.0	0	2,227	30.0 s
Sustained decode	16K	2	Valid	127.9	63.9	358 ms	15.6 ms	2.0	0	3,834	30.0 s
Sustained decode	16K	4	Valid	199.9	49.9	533 ms	20.1 ms	4.0	0	5,995	30.0 s
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	65.7	65.9	432 ms	15.2 ms	1.0	0	1,970	30.0 s
Sustained decode	32K	2	Valid	104.8	52.4	593 ms	19.1 ms	2.0	0	3,144	30.0 s
Sustained decode	32K	4	Valid	148.8	36.9	893 ms	27.1 ms	4.0	0	4,463	30.0 s
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	53.4	53.6	489 ms	18.7 ms	1.0	0	1,603	30.0 s
Sustained decode	64K	2	Valid	76.4	38.2	673 ms	26.2 ms	2.0	0	2,290	30.0 s
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	128K	1	Valid	38.8	39.0	852 ms	25.7 ms	1.0	0	1,164	30.0 s
Sustained decode	128K	2	Over capacity	—	—	—	—	—	—	—	—
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	176K	1	Valid	32.0	32.2	992 ms	31.1 ms	1.0	0	961	30.0 s
Sustained decode	176K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	176K	128	Over capacity	—	—	—	—	—	—	—	—

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