← Back to community benchmarks
gemma-3-12b-it-qat
Performance
4k
tokens
19.5
PP tok/s
10.2
TG tok/s
210178
TTFT (ms)
9.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 127.8 | 7.9 | 8.3 GB | view |
| 4k | 19.5 | 10.2 | 9.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 7.9 | 1.00× |
| 2× | 16.1 | 2.04× |
| 4× | 28.9 | 3.66× |