Back to search results
gemma-4-E2B-it · Q3_K_S
Apple M2 Ultrallama.cpp v0.44L2 Cross-framework verified
This page aggregates 1 real-world runs of gemma-4-E2B-it (Q3_K_S) on Apple M2 Ultra with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
222.24 tok/s
Decode
Decode speed
799.86 tok/s
Prefill
Prefill speed
1.45 s
TTFT
Time to first token
4.83 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
2 days ago
Last verified
Performance
- llama.cpp · IQ2_XS Decode 255.2 · Prefill 1294.22 ·
- llama.cpp · Q3_K_S (current)Decode 222.24 · Prefill 799.86 ·
- Strata · Q3_K_L Decode 291.87 · Prefill 1161.97 ·
Core figures
- Decode (avg)
- 222.24 tok/s
- Prefill (avg)
- 799.86 tok/s
- TTFT (avg)
- 1.45 s
- VRAM (avg)
- 4.83 GB
- MTP acceptance rate
- —
- TTFB
- 1.78 GB
- Power draw
- 23.8 W
Configuration
Member-level fields are taken from the most recent run
- Model
- gemma-4-E2B-it
- Quantization
- Q3_K_S
- Framework
- llama.cpp
- Version
- v0.44
- Context length
- 4096 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 M2 Ultra
- Measured VRAM (avg)
- 4.83 GB
- OS
- macOS 15.0
- Driver
- Apple M2 Ultra
- CUDA
- N/A
- Power draw
- 23.8 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.44 · macOS 15.0 · CUDA N/A · 4096 ctx
222.24 tok/s
Decode
799.86 tok/s
Prefill
1.45 s
TTFT
4.83 GB
VRAM
—
MTP
23.8 W
Power draw
L2 batch seed 2026-10-16: llama.cpp on hw_id=103 running model_id=66 at Q3_K_S