← Back to community benchmarks
Nanbeige4.1-3B
Performance
1k
tokens
1,500
PP tok/s
102.9
TG tok/s
684
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.7
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,500 | 102.9 | 4.6 GB | current |
| 4k | 1,717 | 92.1 | 4.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 102.9 | 1.00× |
| 2× | 190.5 | 1.85× |
| 4× | 266.4 | 2.59× |
| 8× | 297.3 | 2.89× |