Speed test matrix

NVIDIA-Nemotron-Nano-12B-v2-AWQ-4bit on RTX 3060 12GB

vllm · 0.32.0 · AWQ 4-bit5 contexts (0–80K) × 8 concurrency levels (1–128) · prefill at 8K, 16K, 32K, 64K, 80K

by @Sandbergllm-inference-bench v0.8.0Measured Oct 31, 2026, 13:51 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: AWQ 4-bit
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 cyankiwi/NVIDIA-Nemotron-Nano-12B-v2-AWQ-4bit --max-model-len 90112 --gpu-memory-utilization 0.94 --max-num-seqs 36 --enable-prefix-caching --trust-remote-code --kv-cache-dtype int4_per_token_head --attention-backend TRITON_ATTN --max-cudagraph-capture-size 16

Prefill

prompt tok/s by context length

8K context950.08,782 ms TTFT

16K context846.019,516 ms TTFT

32K context716.045,889 ms TTFT

64K context548.0119,763 ms TTFT

80K context489.0167,672 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	40.3140 ms TTFT	76.9216 ms TTFT	141.8403 ms TTFT	189.0Requests queued580 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity
16K	37.22,605 ms TTFT	66.83,901 ms TTFT	113.76,432 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
32K	34.56,372 ms TTFT	59.1Requests queued9,509 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
64K	25.4Requests queued5,461 ms TTFT	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity	Over capacity
80K	18.3Requests queued11,382 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	950.0	—	8,782 ms	—	—	1	—	—
Prefill	16K	1	Valid	846.0	—	19,516 ms	—	—	1	—	—
Prefill	32K	1	Valid	716.0	—	45,889 ms	—	—	1	—	—
Prefill	64K	1	Valid	548.0	—	119,763 ms	—	—	1	—	—
Prefill	80K	1	Valid	489.0	—	167,672 ms	—	—	1	—	—
Sustained decode	0	1	Valid	40.3	40.3	140 ms	24.8 ms	1.0	0	1,208	30.0 s
Sustained decode	0	2	Valid	76.9	38.5	216 ms	26.0 ms	2.0	0	2,305	30.0 s
Sustained decode	0	4	Valid	141.8	35.5	403 ms	28.2 ms	4.0	0	4,252	30.0 s
Sustained decode	0	8	Requests queued	189.0	32.2	580 ms	31.1 ms	6.0	0	5,670	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	37.2	37.2	2,605 ms	26.9 ms	1.0	0	1,115	30.0 s
Sustained decode	16K	2	Valid	66.8	32.3	3,901 ms	31.0 ms	2.0	0	2,004	30.0 s
Sustained decode	16K	4	Valid	113.7	25.7	6,432 ms	39.0 ms	4.0	0	3,408	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	34.5	34.6	6,372 ms	28.9 ms	1.0	0	1,035	30.0 s
Sustained decode	32K	2	Requests queued	59.1	27.2	9,509 ms	37.1 ms	2.0	0	1,774	30.0 s
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	64K	1	Requests queued	25.4	30.4	5,461 ms	32.9 ms	0.8	0	762	30.0 s
Sustained decode	64K	2	Over capacity	—	—	—	—	—	—	—	—
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	80K	1	Requests queued	18.3	28.7	11,382 ms	34.8 ms	0.8	0	549	30.0 s
Sustained decode	80K	2	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	4	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	8	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	16	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	32	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	64	Over capacity	—	—	—	—	—	—	—	—
Sustained decode	80K	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.