Compare AI products
Add up to 4 products and see features, pricing, and scores side by side.
|
Amazon Web Services
|
|
|---|---|
| Score | 8.5/10 |
| Type | hardware |
| Category | Data Center AI Accelerators |
| Summary | Amazon's newest 3nm-class training chip, up to 4.4x faster than Trn2. |
| Starting price | Varies per hour |
| Visit → |
Feature matrix
8 features| Feature | AWS Trainium3 (Trn3) |
|---|---|
| 2.52 PFLOPS FP8 compute per chip | ✓ |
| 144 GB HBM3e per chip, 4.9 TB/s bandwidth | ✓ |
| Trn3 UltraServers scale to 144 chips (362 PFLOPS FP8) | ✓ |
| Up to 20.7 TB HBM3e per UltraServer | ✓ |
| 4x better perf/watt vs Trn2 UltraServers | ✓ |
| EC2 UltraClusters 3.0 for hundreds of thousands of chips | ✓ |
| AWS Neuron SDK (PyTorch/JAX) | ✓ |
| 706 TB/s aggregate memory bandwidth per UltraServer | ✓ |
✓ = the product lists this capability. Capabilities are normalized to a shared vocabulary so the same feature lines up across products.