Arao AI
Back to search results

Qwen3-Coder-Next · Q8_0

Start a discussion
AMD Strix Halo (Radeon 8060S)llama.cpp v0.42L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-Coder-Next (Q8_0) on AMD Strix Halo (Radeon 8060S) with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

46.29 tok/s

Decode

Decode speed

157.5 tok/s

Prefill

Prefill speed

1.17 s

TTFT

Time to first token

1.05 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. llama.cpp · Q8_0 (current)Decode 46.29 · Prefill 157.5 ·
  2. Ollama · Q2_K Decode 85.26 · Prefill 425.23 ·

Core figures

Decode (avg)
46.29 tok/s
Prefill (avg)
157.5 tok/s
TTFT (avg)
1.17 s
VRAM (avg)
1.05 GB
MTP acceptance rate
—
TTFB
0.37 GB
Power draw
71.8 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-Coder-Next
Quantization
Q8_0
Framework
llama.cpp
Version
v0.42
Context length
32768 tokens
Batch size
256
GPU layers
20
Flash Attention
On

Hardware

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

GPU
AMD Strix Halo (Radeon 8060S)
Measured VRAM (avg)
1.05 GB
OS
Ubuntu 22.04
Driver
amdgpu 6.11
CUDA
N/A
Power draw
71.8 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-16

    v0.42 · Ubuntu 22.04 · CUDA N/A · 32768 ctx

    46.29 tok/s

    Decode

    157.5 tok/s

    Prefill

    1.17 s

    TTFT

    1.05 GB

    VRAM

    —

    MTP

    71.8 W

    Power draw

    L2 batch seed 2026-10-16: llama.cpp on hw_id=87 running model_id=58 at Q8_0