← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,143
PP tok/s
57.4
TG tok/s
14341
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.7rc2
macOS
macOS 26.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 959.4 | 68.8 | 35.4 GB | view |
| 4k | 1,390 | 65.2 | 36.2 GB | view |
| 8k | 1,295 | 63.1 | 36.5 GB | view |
| 16k | 1,143 | 57.4 | 37.1 GB | current |
| 32k | 911.6 | 50.3 | 38.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 68.8 | 1.00× |
| 2× | 107.6 | 1.56× |
| 4× | 159.6 | 2.32× |