← Back to community benchmarks
Qwen3-4b-Instruct-2507
Performance
8k
tokens
234.7
PP tok/s
21.9
TG tok/s
34905
TTFT (ms)
5.7
Peak mem (GB)
Hardware
Chip
M1 Pro (16c)
Memory
32 GB
GPU Cores
16
Software
oMLX
v0.2.7
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 277.1 | 34.1 | 4.7 GB | view |
| 4k | 256.7 | 27.1 | 5.3 GB | view |
| 8k | 234.7 | 21.9 | 5.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 34.1 | 1.00× |
| 2× | 56.3 | 1.65× |
| 4× | 62.9 | 1.84× |
| 8× | 56.6 | 1.66× |