← Back to community benchmarks
gemma-4-e4b-it-oQ4e
Performance
32k
tokens
1,009
PP tok/s
44.0
TG tok/s
32482
TTFT (ms)
5.8
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.5.1
macOS
macOS 26.5.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,495 | 66.3 | 4.8 GB | view |
| 4k | 1,292 | 64.1 | 4.9 GB | view |
| 8k | 1,193 | 59.8 | 5.1 GB | view |
| 16k | 1,112 | 54.3 | 5.1 GB | view |
| 32k | 1,009 | 44.0 | 5.8 GB | current |
| 64k | 876.2 | 29.3 | 6.3 GB | view |
| 128k | 685.1 | 18.3 | 8.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 66.3 | 1.00× |
| 2× | 107.3 | 1.62× |