← Back to community benchmarks
Nanbeige4.1-3B
Performance
4k
tokens
2,739
PP tok/s
103.7
TG tok/s
1496
TTFT (ms)
4.8
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,026 | 112.4 | 4.6 GB | view |
| 4k | 2,739 | 103.7 | 4.8 GB | current |
| 8k | 2,689 | 88.9 | 5.1 GB | view |
| 16k | 2,259 | 71.3 | 5.5 GB | view |
| 32k | 1,775 | 51.4 | 6.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 112.4 | 1.00× |
| 2× | 196.6 | 1.75× |
| 4× | 298.9 | 2.66× |