Specifications
Power
| Power Envelope | 10–30 W |
|---|
Memory
| Memory | 64 GB |
|---|---|
| Memory Type | LPDDR4x |
| Memory Bandwidth (GB/s) | 136.5 |
Compute
| CPU Cores | 8 |
|---|---|
| Tensor Cores | 64 |
| NVDLA Engines | 2 |
| GPU CUDA Cores | 512 |
| GPU Architecture | Volta |
Thermal
| Module Max TTP Temperature | 80 °C |
|---|
Identity
| Device Type | System-on-Module |
|---|---|
| Jetson Family | Jetson AGX Xavier |
| Lifecycle Status | EOL |
Mechanical
| Form Factor | 699-pin B2B (100x87mm) |
|---|
Performance
| Max AI Performance (TOPS) | 32 TOPS |
|---|---|
| Max FP32 TFLOPS | 1.41 TFLOPS |
Category: NVIDIA / NVIDIA Hardware / Production Modules
Questions about this product
- Aetina AGX Xavier systems that are end-of-life
- Jetson AGX Xavier 64GB vs Jetson AGX Orin 64GB: memory bandwidth, type and lifecycle
- Jetson modules with 2 NVDLA engines and a max power envelope of 25 W or less
- Jetson modules with more than two NVDLA engines
- Jetson production modules already marked EOL in a supply-chain review
- Jetson production modules with 64 or more Tensor Cores
- Jetson production modules with a Volta GPU
- Jetson production modules with more than 128 GB of memory
- Jetson production modules with the 699-pin B2B (100x87mm) connector
- Lowest-power NVIDIA module with at least 64 GB of memory
- Module with 2 NVDLA engines giving the most AI TOPS at 30 W max or less