← Back to community benchmarks
Qwen3.6-35B-A3B-UD
Performance
1k
tokens
1,101
PP tok/s
75.6
TG tok/s
930
TTFT (ms)
39.4
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8rc1
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,101 | 75.6 | 39.4 GB | current |
| 4k | 1,434 | 72.2 | 40.2 GB | view |
| 8k | 1,449 | 68.0 | 40.5 GB | view |
| 16k | 1,348 | 58.0 | 41.2 GB | view |
| 32k | 1,011 | 43.5 | 42.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 75.6 | 1.00× |
| 2× | 125.7 | 1.66× |
| 4× | 178.3 | 2.36× |
| 8× | 226.9 | 3.00× |