← Back to community benchmarks
gemma-4-e4b-it
Performance
32k
tokens
1,953
PP tok/s
29.7
TG tok/s
16778
TTFT (ms)
16.4
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,662 | 34.6 | 15.3 GB | view |
| 4k | 2,110 | 34.0 | 15.4 GB | view |
| 8k | 2,193 | 33.0 | 15.6 GB | view |
| 16k | 2,111 | 31.7 | 15.8 GB | view |
| 32k | 1,953 | 29.7 | 16.4 GB | current |
| 64k | 1,643 | 25.4 | 17.5 GB | view |
| 128k | 1,326 | 20.1 | 19.7 GB | view |
| 195k | 1,031 | 16.0 | 22.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 34.6 | 1.00× |
| 2× | 34.2 | 0.99× |
| 4× | 66.2 | 1.91× |
| 8× | 128.0 | 3.70× |