← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
4k
tokens
458.5
PP tok/s
45.9
TG tok/s
8934
TTFT (ms)
20.2
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 446.1 | 48.4 | 20.0 GB | view |
| 4k | 458.5 | 45.9 | 20.2 GB | current |
| 8k | 440.3 | 42.1 | 20.4 GB | view |
| 16k | 393.3 | 35.0 | 20.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 48.4 | 1.00× |
| 2× | 74.4 | 1.54× |
| 4× | 86.2 | 1.78× |