← Back to community benchmarks
Qwen3-14B
Performance
1k
tokens
117.1
PP tok/s
12.7
TG tok/s
8748
TTFT (ms)
8.0
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 117.1 | 12.7 | 8.0 GB | current |
| 4k | 112.8 | 9.1 | 9.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.7 | 1.00× |
| 2× | 23.7 | 1.87× |
| 4× | 28.8 | 2.27× |