← Back to community benchmarks
gemma-4-26B-A4B-it-heretic-msq-2.6bit
Performance
4k
tokens
375.4
PP tok/s
29.5
TG tok/s
10910
TTFT (ms)
9.6
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 362.5 | 32.2 | 9.1 GB | view |
| 4k | 375.4 | 29.5 | 9.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.2 | 1.00× |
| 2× | 44.4 | 1.38× |
| 4× | 50.7 | 1.57× |