Arao AI
Back to search results

Llama-3.2-3B-Instruct · IQ3_S

Start a discussion
AMD Radeon RX 7900 XTX24GBvLLM v0.43L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.2-3B-Instruct (IQ3_S) on AMD Radeon RX 7900 XTX with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

253.22 tok/s

Decode

Decode speed

1,074.86 tok/s

Prefill

Prefill speed

1.24 s

TTFT

Time to first token

6.26 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · IQ3_S (current)Decode 253.22 · Prefill 1074.86 ·
  2. Strata · IQ3_S Decode 389.24 · Prefill 2015.9 ·
  3. Strata · Q4_K_S Decode 341.32 · Prefill 1578.68 ·

Core figures

Decode (avg)
253.22 tok/s
Prefill (avg)
1,074.86 tok/s
TTFT (avg)
1.24 s
VRAM (avg)
6.26 GB
MTP acceptance rate
—
TTFB
2.03 GB
Power draw
201 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.2-3B-Instruct
Quantization
IQ3_S
Framework
vLLM
Version
v0.43
Context length
32768 tokens
Batch size
8
GPU layers
999
Flash Attention
On

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
AMD Radeon RX 7900 XTX
Nominal VRAM
24 GB
Measured VRAM (avg)
6.26 GB
OS
Ubuntu 22.04
Driver
amdgpu 6.11
CUDA
12.4
Power draw
201 W

Sources & evidence

1 measured records in total, each traceable to its original source

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.43 · Ubuntu 22.04 · CUDA 12.4 · 32768 ctx

    253.22 tok/s

    Decode

    1,074.86 tok/s

    Prefill

    1.24 s

    TTFT

    6.26 GB

    VRAM

    —

    MTP

    201 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=11 running model_id=70 at IQ3_S