← Back to community benchmarks
Qwen3.5-4B
Performance
8k
tokens
391.9
PP tok/s
31.1
TG tok/s
20902
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.8
macOS
macOS 15.6.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 406.7 | 37.5 | 3.9 GB | view |
| 4k | 403.8 | 34.5 | 4.2 GB | view |
| 8k | 391.9 | 31.1 | 4.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 37.5 | 1.00× |
| 2× | 71.0 | 1.89× |
| 4× | 88.9 | 2.37× |
| 8× | 88.4 | 2.36× |