← All hardware
Amazon's newest 3nm-class training chip, up to 4.4x faster than Trn2.
Pros
- 4.4x performance and 4x perf/watt vs Trn2
- Strong cost efficiency for AWS-native training
- Massive UltraCluster scale-out
Cons
- AWS-only and Neuron-SDK dependent
- Mostly offered via Capacity Blocks / contracts, not flat on-demand
- Smaller third-party ecosystem than CUDA
✓ Where it shines / best for
- AWS-native teams scaling next-gen model training
- Cost/efficiency-sensitive large training and inference jobs
- UltraServer/UltraCluster frontier workloads
✕ Not the best fit for
- Users needing available-now capacity (preview)
- On-premises or edge use
- Stacks that cannot run on the Neuron SDK
Features
- ✓ AI inference
- ✓ Data-center scale
- ✓ LLM
- ✓ API access
- ✓ AI Training
- ✓ Cloud Only
- ✓ Energy Efficient
- ✓ Neuron SDK
- ✓ Next Gen
- ✓ Preview
- ✓ 3nm
Pricing
| Plan | Price | Billing | Notes |
|---|---|---|---|
| On-demand (cloud) | Not yet published | hourly | Trn3 (Trainium3) instances announced in preview; will be priced per instance-hour on EC2 once generally available. No public rates yet; cloud-only, not sold as silicon. |
Pricing verified from the official source. Prices change often — confirm on the vendor's site before buying.
Specifications
| architecture | Trainium3, AWS custom ASIC |
| fp8_per_chip | 2.52 PFLOPS |
| perf_per_watt | 4x vs Trn2 UltraServers |
| memory_per_chip | 144 GB HBM3e, 4.9 TB/s |
| ultraserver_scale | 144 chips, 362 PFLOPS FP8, 20.7 TB HBM3e |
Sponsored
A full review is being generated for this product and will appear here shortly.