← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
1,227
PP tok/s
63.7
TG tok/s
13348
TTFT (ms)
20.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,652 | 93.4 | 20.1 GB | view |
| 4k | 1,706 | 88.6 | 20.2 GB | view |
| 8k | 1,487 | 90.0 | 20.4 GB | view |
| 16k | 1,227 | 63.7 | 20.8 GB | current |
| 32k | 1,026 | 42.5 | 21.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 93.4 | 1.00× |
| 2× | 163.9 | 1.75× |