← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
817.1
PP tok/s
46.9
TG tok/s
20053
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 766.9 | 56.7 | 35.4 GB | view |
| 4k | 898.4 | 54.7 | 36.2 GB | view |
| 8k | 880.9 | 52.1 | 36.5 GB | view |
| 16k | 817.1 | 46.9 | 37.1 GB | current |
| 32k | 676.1 | 39.9 | 38.5 GB | view |
| 64k | 529.0 | 29.6 | 41.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 56.7 | 1.00× |
| 2× | 86.5 | 1.53× |
| 4× | 112.6 | 1.99× |
| 8× | 139.9 | 2.47× |