Arao AI
Back to search results

gemma-4-12B-it · Q4_K_S

Start a discussion
AMD Radeon RX 7900 XTX24GBvLLM v0.27L2 Cross-framework verified

This page aggregates 1 real-world runs of gemma-4-12B-it (Q4_K_S) on AMD Radeon RX 7900 XTX with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

86.73 tok/s

Decode

Decode speed

462.81 tok/s

Prefill

Prefill speed

1.15 s

TTFT

Time to first token

21.76 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · Q4_K_S (current)Decode 86.73 · Prefill 462.81 ·
  2. vLLM · Q5_K_S Decode 64.98 · Prefill 291.13 ·
  3. Ollama · IQ2_M Decode 92.49 · Prefill 473.64 ·
  4. Ollama · Q4_K_S Decode 67.64 · Prefill 313.4 ·
  5. Strata · IQ3_XS Decode 129.13 · Prefill 417.63 ·

Core figures

Decode (avg)
86.73 tok/s
Prefill (avg)
462.81 tok/s
TTFT (avg)
1.15 s
VRAM (avg)
21.76 GB
MTP acceptance rate
—
TTFB
0.69 GB
Power draw
381.1 W

Configuration

Member-level fields are taken from the most recent run

Model
gemma-4-12B-it
Quantization
Q4_K_S
Framework
vLLM
Version
v0.27
Context length
32768 tokens
Batch size
16
GPU layers
999
Flash Attention
Off

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)
21.76 GB
OS
macOS 15.0
Driver
565.57
CUDA
12.6
Power draw
381.1 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

    v0.27 · macOS 15.0 · CUDA 12.6 · 32768 ctx

    86.73 tok/s

    Decode

    462.81 tok/s

    Prefill

    1.15 s

    TTFT

    21.76 GB

    VRAM

    —

    MTP

    381.1 W

    Power draw

    L2 batch seed 2026-10-10: vLLM on hw_id=11 running model_id=65 at Q4_K_S