Back to search results
gemma-4-12B-it · Q2_K
Apple M2 UltraOllama v0.23L2 Cross-framework verified
This page aggregates 1 real-world runs of gemma-4-12B-it (Q2_K) on Apple M2 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.
96.52 tok/s
Decode
Decode speed
478.87 tok/s
Prefill
Prefill speed
1.57 s
TTFT
Time to first token
15.96 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- llama.cpp · IQ4_XS Decode 90.66 · Prefill 438.77 ·
- llama.cpp · Q5_K_S Decode 66.9 · Prefill 326.75 ·
- Ollama · Q2_K (current)Decode 96.52 · Prefill 478.87 ·
Core figures
- Decode (avg)
- 96.52 tok/s
- Prefill (avg)
- 478.87 tok/s
- TTFT (avg)
- 1.57 s
- VRAM (avg)
- 15.96 GB
- MTP acceptance rate
- —
- TTFB
- 0.77 GB
- Power draw
- 131 W
Configuration
Member-level fields are taken from the most recent run
- Model
- gemma-4-12B-it
- Quantization
- Q2_K
- Framework
- Ollama
- Version
- v0.23
- Context length
- 4096 tokens
- Batch size
- 512
- 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
- Apple M2 Ultra
- Measured VRAM (avg)
- 15.96 GB
- OS
- macOS 26.1
- Power draw
- 131 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.23 · macOS 26.1 · 4096 ctx
96.52 tok/s
Decode
478.87 tok/s
Prefill
1.57 s
TTFT
15.96 GB
VRAM
—
MTP
131 W
Power draw
L2 batch seed 2026-10-19: Ollama on hw_id=103 running model_id=65 at Q2_K