Back to search results
gemma-4-12B-it · Q5_K_S
Apple M4 Pro64GBllama.cpp v0.26L2 Cross-framework verified
This page aggregates 1 real-world runs of gemma-4-12B-it (Q5_K_S) on Apple M4 Pro with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
68.46 tok/s
Decode
Decode speed
358.45 tok/s
Prefill
Prefill speed
1.58 s
TTFT
Time to first token
24.78 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
5 days ago
Last verified
Performance
- llama.cpp · IQ2_M Decode 103.7 · Prefill 452.45 ·
- llama.cpp · Q5_K_S (current)Decode 68.46 · Prefill 358.45 ·
- vLLM · Q2_K Decode 110.03 · Prefill 407.67 ·
- vLLM · Q4_K_S Decode 90.85 · Prefill 484.54 ·
Core figures
- Decode (avg)
- 68.46 tok/s
- Prefill (avg)
- 358.45 tok/s
- TTFT (avg)
- 1.58 s
- VRAM (avg)
- 24.78 GB
- MTP acceptance rate
- —
- TTFB
- 0.55 GB
- Power draw
- 42.7 W
Configuration
Member-level fields are taken from the most recent run
- Model
- gemma-4-12B-it
- Quantization
- Q5_K_S
- Framework
- llama.cpp
- Version
- v0.26
- Context length
- 32768 tokens
- Batch size
- 512
- GPU layers
- 33
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- Apple M4 Pro
- Nominal VRAM
- 64 GB
- Measured VRAM (avg)
- 24.78 GB
- OS
- macOS 15.0
- Driver
- Apple M4 Pro
- CUDA
- N/A
- Power draw
- 42.7 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.26 · macOS 15.0 · CUDA N/A · 32768 ctx
68.46 tok/s
Decode
358.45 tok/s
Prefill
1.58 s
TTFT
24.78 GB
VRAM
—
MTP
42.7 W
Power draw
L2 batch seed 2026-10-13: llama.cpp on hw_id=25 running model_id=65 at Q5_K_S