← Back to community benchmarks
Qwen3-4B
Performance
4k
tokens
260.3
PP tok/s
36.5
TG tok/s
15736
TTFT (ms)
3.6
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 266.4 | 45.4 | 2.9 GB | view |
| 4k | 260.3 | 36.5 | 3.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 45.4 | 1.00× |
| 2× | 52.4 | 1.15× |
| 4× | 57.5 | 1.27× |