Full view of a parts catalogue asset.
| Specifications | Value |
|---|---|
| vendor | nvidia |
| manufacturer | NVIDIA |
| gpu name | GA10B |
| architecture | Ampere |
| generation | Tegra(Ampere) |
| release date | 2023-03-01 |
| memory size gb | 8 |
| memory type | LPDDR5 |
| memory bus bits | 128 |
| memory bandwidth gb s | 102.4 |
| tdp w | 20 |
| length mm | 70 |
| slot | IGP |
| fp32 tflops | 1.57 |
| Field | Value |
|---|---|
| foundry | Samsung |
| processSize | 8 |
| dieSize | 200 |
| busInterface | PCIe 4.0 x4 |
| baseClock | 765 |
| boostClock | 765 |
| memoryClock | 800 |
| shaders | 1024 |
| tmus | 32 |
| rops | 16 |
| sms | 8 |
| tensorCores | 32 |
| l1Cache | 128 |
| l2Cache | 4 |
| pixelRate | 12.24 |
| textureRate | 24.48 |
| fp16 | 3.13 |
| directX | 12.2 |
| openGL | 4.6 |
| vulkan | 1.4 |
| openCL | 3.0 |
| cuda | 8.7 |
| shaderModel | 6.8 |
| warpSize | 32 |
| maxThreadsPerBlock | 1024 |
| maxThreadsPerSM | 1536 |
| maxBlocksPerSM | 16 |
| sharedMemPerSM | 167936 |
| registersPerSM | 65536 |
| coresPerSM | 128 |
| Market Item | Status | Match | Confidence | Link Status |
|---|---|---|---|---|
| CMP 170HX 8 GB | observed | exact | 87% | proposed |
| GeForce RTX 3060 Ti | observed | exact | 87% | proposed |
| Quadro RTX 4000 | observed | exact | 87% | proposed |
| Radeon RX 5700 XT | observed | exact | 87% | proposed |
| Radeon RX 6600 | observed | exact | 87% | proposed |
| Source | rightnow_gpu_database |
| Source Asset ID | nvidia-jetson-orin-nx-8-gb |
| Model | Jetson Orin NX 8 GB |
| Benchmark | — |
| Created | 05/06/2026 00:25:55 |
| Updated | 05/06/2026 00:25:55 |