← Back to community benchmarks
Qwen3.6-27B-uncensored-heretic-v2-Native-MTP-Preserved
Performance
4k
tokens
703.1
PP tok/s
9.0
TG tok/s
5826
TTFT (ms)
54.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 510.3 | 9.2 | 52.7 GB | view |
| 4k | 703.1 | 9.0 | 54.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 9.2 | 1.00× |
| 2× | 9.9 | 1.08× |
| 4× | 18.3 | 1.99× |