← Back to community benchmarks
Qwen3.6-35B-A3B-oQ6-mtp
Performance
16k
tokens
553.9
PP tok/s
42.3
TG tok/s
29579
TTFT (ms)
30.2
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.9.dev1
macOS
macOS 15.7.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 461.9 | 50.5 | 28.3 GB | view |
| 4k | 569.2 | 48.1 | 29.0 GB | view |
| 8k | 572.8 | 46.7 | 29.5 GB | view |
| 16k | 553.9 | 42.3 | 30.2 GB | current |
| 32k | 500.7 | 38.4 | 31.7 GB | view |
| 64k | 410.5 | 30.6 | 34.7 GB | view |
| 128k | 274.2 | 22.2 | 40.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 50.5 | 1.00× |
| 2× | 60.2 | 1.19× |
| 4× | 94.3 | 1.87× |