← Back to community benchmarks
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-oQ6-fp16-mtp
Performance
64k
tokens
926.3
PP tok/s
36.5
TG tok/s
70749
TTFT (ms)
33.6
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.4.2rc1
macOS
macOS 15.7.7
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,354 | 63.1 | 27.8 GB | view |
| 4k | 1,835 | 58.5 | 28.6 GB | view |
| 8k | 1,765 | 58.6 | 28.9 GB | view |
| 16k | 1,594 | 54.8 | 29.6 GB | view |
| 32k | 1,377 | 50.3 | 30.9 GB | view |
| 64k | 926.3 | 36.5 | 33.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 63.1 | 1.00× |
| 2× | 57.1 | 0.90× |
| 4× | 122.9 | 1.95× |