← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
578.3
PP tok/s
55.4
TG tok/s
14166
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 549.1 | 61.6 | 20.1 GB | view |
| 4k | 585.5 | 59.3 | 20.2 GB | view |
| 8k | 578.3 | 55.4 | 20.4 GB | current |
| 16k | 550.6 | 28.0 | 20.8 GB | view |
| 32k | 502.2 | 14.9 | 21.6 GB | view |
| 64k | 404.7 | 7.7 | 23.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 61.6 | 1.00× |
| 2× | 92.6 | 1.50× |
| 4× | 117.3 | 1.90× |