← Back to community benchmarks
Qwen3.6-35B-A3B-oQ6-fp16-mtp
Performance
8k
tokens
1,918
PP tok/s
77.8
TG tok/s
4272
TTFT (ms)
29.6
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,327 | 81.4 | 28.3 GB | view |
| 4k | 1,855 | 80.2 | 29.1 GB | view |
| 8k | 1,918 | 77.8 | 29.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 81.4 | 1.00× |
| 2× | 125.3 | 1.54× |
| 4× | 180.4 | 2.22× |
| 8× | 251.8 | 3.09× |