← Back to community benchmarks
Nex-N2-Pro5.3bit-vision
Performance
128k
tokens
287.7
PP tok/s
20.7
TG tok/s
455520
TTFT (ms)
255.0
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.5.1
macOS
macOS 26.5.2
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 482.6 | 30.0 | 248.7 GB | view |
| 4k | 546.6 | 29.6 | 250.3 GB | view |
| 8k | 536.4 | 29.0 | 250.8 GB | view |
| 16k | 509.3 | 27.9 | 251.1 GB | view |
| 32k | 459.3 | 26.6 | 251.6 GB | view |
| 64k | 383.6 | 24.2 | 253.2 GB | view |
| 128k | 287.7 | 20.7 | 255.0 GB | current |
| 195k | 223.4 | 17.8 | 257.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 30.0 | 1.00× |
| 2× | 46.4 | 1.55× |
| 4× | 69.9 | 2.33× |