Arao AI
Back to search results

Qwen3.5-0.8B · Q3_K_M

Start a discussion
AMD Radeon RX 7900 XTX24GBllama.cpp v0.48L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-0.8B (Q3_K_M) on AMD Radeon RX 7900 XTX with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

251.67 tok/s

Decode

Decode speed

1,158.4 tok/s

Prefill

Prefill speed

1.07 s

TTFT

Time to first token

3.24 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. llama.cpp · Q3_K_M (current)Decode 251.67 · Prefill 1158.4 ·
  2. vLLM · IQ3_XS Decode 312.18 · Prefill 1202.88 ·
  3. vLLM · Q6_K Decode 198.86 · Prefill 984.35 ·

Core figures

Decode (avg)
251.67 tok/s
Prefill (avg)
1,158.4 tok/s
TTFT (avg)
1.07 s
VRAM (avg)
3.24 GB
MTP acceptance rate
—
TTFB
2.01 GB
Power draw
356.1 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-0.8B
Quantization
Q3_K_M
Framework
llama.cpp
Version
v0.48
Context length
4096 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 Radeon RX 7900 XTX
Nominal VRAM
24 GB
Measured VRAM (avg)
3.24 GB
OS
macOS 15.0
Driver
560.94
CUDA
12.4
Power draw
356.1 W

Sources & evidence

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

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

    v0.48 · macOS 15.0 · CUDA 12.4 · 4096 ctx

    251.67 tok/s

    Decode

    1,158.4 tok/s

    Prefill

    1.07 s

    TTFT

    3.24 GB

    VRAM

    —

    MTP

    356.1 W

    Power draw

    L2 batch seed 2026-10-10: llama.cpp on hw_id=11 running model_id=51 at Q3_K_M