← Back to community benchmarks

translategemma-4b-it

M3 Ultra (80c) · 512 GB · unknown · 2026-03-11
Performance
4k
tokens
2,341
PP tok/s
52.2
TG tok/s
1750
TTFT (ms)
8.9
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,850 33.0 8.7 GB view
4k 2,341 52.2 8.9 GB current
8k 2,543 57.4 9.1 GB view
Batching Results
Batch Size TG tok/s Speedup
33.0 1.00×
36.2 1.10×
60.1 1.82×