← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
32k
tokens
513.1
PP tok/s
44.4
TG tok/s
63857
TTFT (ms)
21.6
Peak mem (GB)
Hardware
Chip
M2 Max (30c)
Memory
64 GB
GPU Cores
30
Software
oMLX
v0.2.7
macOS
macOS 15.7.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 555.6 | 75.8 | 20.1 GB | view |
| 4k | 598.0 | 72.7 | 20.2 GB | view |
| 8k | 589.8 | 67.6 | 20.4 GB | view |
| 16k | 563.3 | 57.1 | 20.8 GB | view |
| 32k | 513.1 | 44.4 | 21.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 75.8 | 1.00× |
| 2× | 151.4 | 2.00× |
| 4× | 301.6 | 3.98× |