← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
1k
tokens
1,673
PP tok/s
91.7
TG tok/s
612
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.7
macOS
macOS 15.3.2
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,673 | 91.7 | 20.0 GB | current |
| 4k | 2,386 | 87.2 | 20.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 91.7 | 1.00× |
| 2× | 164.8 | 1.80× |
| 4× | 270.4 | 2.95× |
| 8× | 420.7 | 4.59× |