← Back to community benchmarks
Qwen3.5-4B
Performance
4k
tokens
283.2
PP tok/s
48.2
TG tok/s
14464
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip
M1 Pro (16c)
Memory
16 GB
GPU Cores
16
Software
oMLX
v0.2.9
macOS
macOS 15.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 283.5 | 54.2 | 4.0 GB | view |
| 4k | 283.2 | 48.2 | 4.2 GB | current |
| 8k | 274.1 | 42.2 | 4.6 GB | view |
| 16k | 265.6 | 37.9 | 5.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.2 | 1.00× |
| 2× | 62.1 | 1.15× |
| 4× | 67.1 | 1.24× |
| 8× | 57.0 | 1.05× |