← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
32k
tokens
524.4
PP tok/s
46.2
TG tok/s
62490
TTFT (ms)
21.6
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 732.0 | 80.5 | 20.1 GB | view |
| 4k | 726.0 | 77.3 | 20.2 GB | view |
| 8k | 632.0 | 64.8 | 20.4 GB | view |
| 16k | 592.5 | 56.8 | 20.8 GB | view |
| 32k | 524.4 | 46.2 | 21.6 GB | current |
| 64k | 433.3 | 32.4 | 23.3 GB | view |
| 128k | 296.9 | 5.0 | 59.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 80.5 | 1.00× |
| 2× | 106.9 | 1.33× |