← Back to community benchmarks
gemma-4-E2B-it-oQ8
Performance
16k
tokens
4,852
PP tok/s
75.1
TG tok/s
3377
TTFT (ms)
6.2
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.4.1
macOS
macOS 15.7.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 3,344 | 82.8 | 6.0 GB | view |
| 4k | 4,803 | 81.7 | 6.1 GB | view |
| 8k | 4,787 | 83.3 | 6.2 GB | view |
| 16k | 4,852 | 75.1 | 6.2 GB | current |
| 32k | 4,363 | 70.0 | 6.6 GB | view |
| 64k | 3,307 | 57.6 | 7.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 82.8 | 1.00× |
| 2× | 152.5 | 1.84× |
| 4× | 240.0 | 2.90× |
| 8× | 487.2 | 5.88× |