← Back to community benchmarks
Qwen3-4B
Performance
1k
tokens
1,337
PP tok/s
138.3
TG tok/s
766
TTFT (ms)
2.9
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.11
macOS
macOS 15.7.3
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,337 | 138.3 | 2.9 GB | current |
| 4k | 1,335 | 113.2 | 3.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 138.3 | 1.00× |
| 2× | 232.1 | 1.68× |
| 4× | 269.6 | 1.95× |
| 8× | 270.3 | 1.95× |