← Back to community benchmarks
Qwen3.6-27B-uncensored-heretic-v2-Native-MTP-Preserved-oQ4
Performance
16k
tokens
219.5
PP tok/s
29.3
TG tok/s
74654
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.4.4rc1
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 210.2 | 32.4 | 16.5 GB | view |
| 4k | 222.1 | 31.5 | 17.9 GB | view |
| 8k | 223.5 | 30.6 | 18.9 GB | view |
| 16k | 219.5 | 29.3 | 20.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.4 | 1.00× |
| 2× | 44.6 | 1.38× |