Arao AI
Back to search results

Llama-3-8B-Instruct · IQ2_M

Start a discussion
AMD Ryzen AI Max+ 395128GBvLLM v0.31L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3-8B-Instruct (IQ2_M) on AMD Ryzen AI Max+ 395 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

48.9 tok/s

Decode

Decode speed

230.21 tok/s

Prefill

Prefill speed

1.68 s

TTFT

Time to first token

10.84 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. vLLM · IQ2_M (current)Decode 48.9 · Prefill 230.21 ·
  2. vLLM · Q5_K_M Decode 31.8 · Prefill 157.2 ·
  3. Ollama · Q3_K_L Decode 37.4 · Prefill 165.99 ·

Core figures

Decode (avg)
48.9 tok/s
Prefill (avg)
230.21 tok/s
TTFT (avg)
1.68 s
VRAM (avg)
10.84 GB
MTP acceptance rate
—
TTFB
0.39 GB
Power draw
56.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3-8B-Instruct
Quantization
IQ2_M
Framework
vLLM
Version
v0.31
Context length
2048 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 Ryzen AI Max+ 395
Nominal VRAM
128 GB
Measured VRAM (avg)
10.84 GB
OS
Ubuntu 24.04
Driver
amdgpu 6.11
CUDA
12.8
Power draw
56.7 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.31 · Ubuntu 24.04 · CUDA 12.8 · 2048 ctx

    48.9 tok/s

    Decode

    230.21 tok/s

    Prefill

    1.68 s

    TTFT

    10.84 GB

    VRAM

    —

    MTP

    56.7 W

    Power draw

    L2 batch seed 2026-10-19: vLLM on hw_id=24 running model_id=69 at IQ2_M