← Back to community benchmarks
Qwen3.6-35B-A3B-oQ6-mtp
Performance
16k
tokens
3,825
PP tok/s
96.1
TG tok/s
4283
TTFT (ms)
30.3
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,702 | 106.9 | 28.3 GB | view |
| 4k | 3,625 | 102.8 | 29.1 GB | view |
| 8k | 3,909 | 97.1 | 29.6 GB | view |
| 16k | 3,825 | 96.1 | 30.3 GB | current |
| 32k | 3,268 | 85.9 | 31.8 GB | view |
| 64k | 2,324 | 73.8 | 34.8 GB | view |
| 128k | 1,368 | 49.2 | 40.8 GB | view |
| 195k | 791.3 | 38.2 | 47.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 106.9 | 1.00× |
| 2× | 200.2 | 1.87× |
| 4× | 379.2 | 3.55× |