← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
1,216
PP tok/s
90.4
TG tok/s
6736
TTFT (ms)
19.6
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
48 GB
GPU Cores
40
Software
oMLX
v0.2.10
macOS
macOS 26.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,158 | 104.0 | 19.2 GB | view |
| 4k | 1,203 | 100.6 | 19.4 GB | view |
| 8k | 1,216 | 90.4 | 19.6 GB | current |
| 16k | 1,188 | 63.3 | 20.0 GB | view |
| 32k | 1,035 | 42.7 | 20.8 GB | view |
| 64k | 729.7 | 25.9 | 22.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 104.0 | 1.00× |
| 2× | 169.0 | 1.62× |
| 4× | 235.8 | 2.27× |