← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
4k
tokens
2,685
PP tok/s
73.4
TG tok/s
1525
TTFT (ms)
36.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,543 | 75.3 | 36.2 GB | view |
| 4k | 2,685 | 73.4 | 36.3 GB | current |
| 8k | 2,612 | 70.7 | 36.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 75.3 | 1.00× |
| 2× | 132.5 | 1.76× |
| 4× | 216.5 | 2.88× |
| 8× | 338.5 | 4.50× |