← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4e-mtp
Performance
16k
tokens
509.1
PP tok/s
51.6
TG tok/s
32180
TTFT (ms)
21.2
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.5.1
macOS
macOS 26.5.2
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 570.8 | 59.1 | 20.0 GB | view |
| 4k | 611.8 | 57.1 | 20.7 GB | view |
| 8k | 578.5 | 54.1 | 20.9 GB | view |
| 16k | 509.1 | 51.6 | 21.2 GB | current |
| 32k | 412.8 | 47.3 | 22.0 GB | view |
| 64k | 300.7 | 39.0 | 22.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 59.1 | 1.00× |
| 2× | 88.6 | 1.50× |
| 4× | 108.0 | 1.83× |