← Back to community benchmarks
gemma-4-26B-A4B-it-uncensored-oQ8
Performance
16k
tokens
3,337
PP tok/s
65.0
TG tok/s
4910
TTFT (ms)
26.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,031 | 86.8 | 25.7 GB | view |
| 4k | 3,097 | 82.0 | 26.2 GB | view |
| 8k | 3,372 | 75.5 | 26.3 GB | view |
| 16k | 3,337 | 65.0 | 26.7 GB | current |
| 32k | 3,005 | 51.0 | 27.5 GB | view |
| 64k | 2,438 | 34.9 | 29.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 86.8 | 1.00× |
| 2× | 131.5 | 1.51× |
| 4× | 180.0 | 2.07× |
| 8× | 347.4 | 4.00× |