← Back to community benchmarks
Qwen2-0.5B-CosyVoice-BlankEN
Performance
1k
tokens
4,171
PP tok/s
89.3
TG tok/s
246
TTFT (ms)
1.4
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.4
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 4,171 | 89.3 | 1.4 GB | current |
| 4k | 3,500 | 79.3 | 1.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 89.3 | 1.00× |
| 2× | 112.0 | 1.25× |
| 4× | 214.7 | 2.40× |