← Back to community benchmarks
Qwen3.6-35B-A3B-oQ8e-mtp
Performance
32k
tokens
1,599
PP tok/s
64.1
TG tok/s
20499
TTFT (ms)
36.8
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.5.3
macOS
macOS 26.5.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,020 | 65.7 | 35.5 GB | view |
| 4k | 2,251 | 50.5 | 36.2 GB | view |
| 8k | 2,115 | 65.6 | 36.3 GB | view |
| 16k | 1,900 | 65.7 | 36.4 GB | view |
| 32k | 1,599 | 64.1 | 36.8 GB | current |
| 64k | 1,208 | 55.9 | 37.6 GB | view |
| 128k | 813.5 | 46.8 | 39.1 GB | view |
| 195k | 602.2 | 39.3 | 40.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 65.7 | 1.00× |
| 2× | 121.0 | 1.84× |
| 4× | 205.4 | 3.13× |