← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,596
PP tok/s
76.0
TG tok/s
10267
TTFT (ms)
37.4
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.1
macOS
macOS 26.5.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,357 | 87.3 | 35.4 GB | view |
| 4k | 1,693 | 84.2 | 36.2 GB | view |
| 8k | 1,697 | 82.4 | 36.7 GB | view |
| 16k | 1,596 | 76.0 | 37.4 GB | current |
| 32k | 1,385 | 70.5 | 38.9 GB | view |
| 64k | 1,066 | 58.8 | 41.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 87.3 | 1.00× |
| 2× | 140.8 | 1.61× |
| 4× | 193.1 | 2.21× |