← Back to community benchmarks
Qwen3.6-35B-A3B-MTP-Holo3-Qwopus
Performance
128k
tokens
863.2
PP tok/s
18.6
TG tok/s
151843
TTFT (ms)
76.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.3
macOS
macOS 27.0
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,872 | 67.0 | 65.8 GB | view |
| 4k | 2,880 | 66.4 | 66.6 GB | view |
| 8k | 3,809 | 65.7 | 66.9 GB | view |
| 16k | 3,245 | 63.6 | 67.6 GB | view |
| 32k | 2,589 | 60.7 | 68.9 GB | view |
| 64k | 1,950 | 52.9 | 71.5 GB | view |
| 128k | 863.2 | 18.6 | 76.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 67.0 | 1.00× |
| 2× | 27.2 | 0.41× |
| 4× | 67.9 | 1.01× |