← Back to community benchmarks

gemma-4-e2b-it

M1 (7c) · 8 GB · 6bit · 2026-04-05
Performance
16k
tokens
651.1
PP tok/s
20.9
TG tok/s
25164
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip M1 (7c)
Memory 8 GB
GPU Cores 7
Software
oMLX v0.3.4
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 563.1 25.7 4.0 GB view
4k 678.5 24.8 4.1 GB view
8k 681.1 22.8 4.1 GB view
16k 651.1 20.9 4.2 GB current
32k 581.3 17.5 4.6 GB view
Batching Results
Batch Size TG tok/s Speedup
25.7 1.00×
35.7 1.39×