← Back to community benchmarks
gemma-4-26B-A4B-it-TurboQuant
Performance
32k
tokens
285.9
PP tok/s
22.4
TG tok/s
114608
TTFT (ms)
16.4
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.8
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 361.2 | 33.4 | 14.2 GB | view |
| 4k | 367.0 | 32.7 | 14.8 GB | view |
| 8k | 346.4 | 30.0 | 15.2 GB | view |
| 16k | 309.5 | 24.8 | 15.5 GB | view |
| 32k | 285.9 | 22.4 | 16.4 GB | current |
| 64k | 232.8 | 17.6 | 18.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 33.4 | 1.00× |
| 2× | 25.7 | 0.77× |