← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
448.9
PP tok/s
42.6
TG tok/s
18249
TTFT (ms)
20.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 15.7.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 425.4 | 47.9 | 19.2 GB | view |
| 4k | 463.5 | 45.7 | 19.9 GB | view |
| 8k | 448.9 | 42.6 | 20.2 GB | current |
| 16k | 412.2 | 34.8 | 20.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.9 | 1.00× |
| 2× | 72.0 | 1.50× |
| 4× | 84.0 | 1.75× |