Beijing Plink AI Technology Co., Ltd
Beijing Plink AI is an expert of cloud-to-end one-stop solution.
Add to Cart
| Module | CUDA Cores | Tensor Cores | Operating Frequency per Core (up to) |
| Jetson Orin Nano 8GB | 1024 | 32 | 625 MHz |
| Jetson Orin Nano 4GB | 512 | 16 | 625 MHz |
| AI Perf | Up to 20 (Sparse) INT8 TOPs and 10 (Dense) INT8 TOPs |
| Ampere GPU | 512 NVIDIA® CUDA® cores | 16 Tensor cores |
| DARM Cortex-A78AE CPU | Six-core (ON 8GB and ON 4GB) Cortex A78AE ARMv8.2 (64-bit) heterogeneous multi-processing (HMP) CPU architecture | 2x clusters (1x 4-core cluster + 128 KB L1 + 256KB L2 per core + 2MB L3) + 1x 2- core cluster (128 KB L1 + 256KB L2 per core + 2MB L3) | System Cache: 4 MB (shared across all clusters) |
| Temp. Range | -25°C – 90°C |
| Supported Power Input | 5V-20V |
| Module Size | 69.6 mm x 45 mm | 260 pin SO-DIMM Connector |
| Memory | 4GB 64-bit LPDDR5 DRAM |
| Networking | 10/100/1000 BASE-T Ethernet | Media Access Controller (MAC) |
| Storage | Supports External Storage (NVMe) |
| Standard | Profile(s) | Resolution (Maximum Number of Streams) | Throughput MPix/s (Up to) | Maximum Cumulative Bitrate (Mbps) |
| H.264 | Baseline, Main, High | 4K30 (1) | 1080p60 (3) | 1080p30 (7) | 450 | 115 |
| High 444, High 444 Predictive, MVC (per view considering two views) * | 1x1080p60 | 3x 1080p30 | 225 | 100 | |
| H.265 (HEVC) | Main, Main10 | 4K60 (1) | 4K30 (2) | 1080p60 (5) | 1080p30 (11) | 690 | 150 |
| Main 444, Main 444 10, MV (per view) | 4K30 (1) | 1080p60 (2) | 1080p30 (5) | 340 | 100 | |
| AV1 | Main Profile | 4K60 (1) | 4K30 (2) | 1080p60 (5) | 1080p30 (10) | 625 | 75 |
| VP9 | Profile 0, Profile 2 | 4K60 (1) | 4K30 (2) | 1080p60 (5) | 1080p30 (10) | 625 | 100 |
• * Maximum throughput half for YUV444 – as compared to YUV420
• AV1 Dec: Maximum aggregate Mbps should be <120
