← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4-mtp
Performance
16k
tokens
581.4
PP tok/s
48.5
TG tok/s
28179
TTFT (ms)
22.1
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
48 GB
GPU Cores
16
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 | 535.8 | 57.7 | 20.1 GB | view |
| 4k | 623.7 | 57.4 | 20.5 GB | view |
| 8k | 643.4 | 55.6 | 21.5 GB | view |
| 16k | 581.4 | 48.5 | 22.1 GB | current |
| 32k | 512.6 | 43.9 | 23.5 GB | view |
| 64k | 406.9 | 36.6 | 26.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 57.7 | 1.00× |
| 2× | 86.6 | 1.50× |
| 4× | 176.4 | 3.06× |