← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
1k
tokens
1,578
PP tok/s
77.3
TG tok/s
649
TTFT (ms)
36.2
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,578 | 77.3 | 36.2 GB | current |
| 4k | 1,714 | 73.3 | 36.3 GB | view |
| 8k | 1,651 | 72.5 | 36.5 GB | view |
| 16k | 1,460 | 62.1 | 37.0 GB | view |
| 32k | 1,156 | 48.7 | 37.8 GB | view |
| 64k | 943.9 | 15.0 | 39.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 77.3 | 1.00× |
| 2× | 137.1 | 1.77× |
| 4× | 197.4 | 2.55× |
| 8× | 243.8 | 3.15× |