← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4-mtp
Performance
16k
tokens
826.0
PP tok/s
65.1
TG tok/s
19836
TTFT (ms)
22.4
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
48 GB
GPU Cores
20
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 787.3 | 76.0 | 20.4 GB | view |
| 4k | 907.8 | 71.2 | 21.2 GB | view |
| 8k | 889.1 | 68.5 | 21.5 GB | view |
| 16k | 826.0 | 65.1 | 22.4 GB | current |
| 32k | 687.8 | 55.4 | 23.9 GB | view |
| 64k | 538.2 | 41.7 | 26.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 76.0 | 1.00× |
| 2× | 112.3 | 1.48× |
| 4× | 142.3 | 1.87× |