← Back to community benchmarks
Qwen3.6-35B-A3B-mxfp8
Performance
64k
tokens
2,213
PP tok/s
63.4
TG tok/s
29615
TTFT (ms)
39.9
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,187 | 73.0 | 35.1 GB | view |
| 4k | 1,473 | 67.9 | 35.9 GB | view |
| 8k | 1,482 | 67.3 | 36.3 GB | view |
| 16k | 3,073 | 70.0 | 37.5 GB | view |
| 32k | 2,817 | 67.3 | 38.0 GB | view |
| 64k | 2,213 | 63.4 | 39.9 GB | current |
| 128k | 1,555 | 55.8 | 42.6 GB | view |
| 195k | 1,036 | 45.9 | 45.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 73.0 | 1.00× |
| 2× | 114.8 | 1.57× |
| 4× | 134.1 | 1.84× |
| 8× | 161.1 | 2.21× |