← Back to community benchmarks
Qwen3.5-4B
Performance
4k
tokens
338.8
PP tok/s
31.5
TG tok/s
12090
TTFT (ms)
4.2
Peak mem (GB)
Hardware
Chip
M3 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 15.7.4
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 337.8 | 33.0 | 3.9 GB | view |
| 4k | 338.8 | 31.5 | 4.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 33.0 | 1.00× |
| 2× | 63.3 | 1.92× |
| 4× | 78.6 | 2.38× |