← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
437.0
PP tok/s
43.1
TG tok/s
37492
TTFT (ms)
21.0
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
32 GB
GPU Cores
24
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 385.8 | 61.0 | 19.3 GB | view |
| 4k | 456.7 | 57.6 | 20.0 GB | view |
| 8k | 456.1 | 53.2 | 20.4 GB | view |
| 16k | 437.0 | 43.1 | 21.0 GB | current |
| 32k | 398.3 | 31.6 | 22.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 61.0 | 1.00× |
| 2× | 80.7 | 1.32× |
| 4× | 96.0 | 1.57× |
| 8× | 108.1 | 1.77× |