← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
64k
tokens
1,294
PP tok/s
52.3
TG tok/s
50650
TTFT (ms)
23.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,101 | 95.8 | 20.1 GB | view |
| 4k | 2,163 | 92.6 | 20.2 GB | view |
| 8k | 2,101 | 87.3 | 20.4 GB | view |
| 16k | 1,941 | 79.8 | 20.8 GB | view |
| 32k | 1,678 | 68.5 | 21.6 GB | view |
| 64k | 1,294 | 52.3 | 23.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 95.8 | 1.00× |
| 2× | 160.1 | 1.67× |
| 4× | 254.7 | 2.66× |
| 8× | 378.5 | 3.95× |