← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
676.2
PP tok/s
63.7
TG tok/s
24229
TTFT (ms)
21.0
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
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 | 658.7 | 83.0 | 19.3 GB | view |
| 4k | 748.6 | 78.3 | 20.0 GB | view |
| 8k | 732.6 | 72.7 | 20.4 GB | view |
| 16k | 676.2 | 63.7 | 21.0 GB | current |
| 32k | 571.1 | 48.0 | 22.4 GB | view |
| 64k | 445.1 | 32.3 | 25.0 GB | view |
| 128k | 292.7 | 19.7 | 30.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.0 | 1.00× |
| 2× | 117.4 | 1.41× |