Supercomputers/Systems/AWS Trainium3 UltraServer
USA · operational · mixed
AWS Trainium3 UltraServer
Also known as Trn3 UltraServer
Operated by Amazon .
Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1540 systems have a readable citation that names them, and 64 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- -
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- -
- Power
- -
- Per watt
- - GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - AWS-designed and operated Trn3 UltraServer in EC2 UltraClusters 3.0 | Amazon | - | Reported | aws.amazon.com |
| Accelerator | AWS Trainium3 Accelerators NeuronCore-v4 Trn3 UltraServer: up to 144 Trainium3 chips on NeuronLink-v4 through NeuronSwitch-v1 | Amazon | 144 reported | Reported | aws.amazon.com |
AI datacenter metrics
Figures beyond an accelerator count, each with who stated it. What can and cannot be compared across AI datacenters.
| Metric | Value | As of | Basis | Note | Source |
|---|---|---|---|---|---|
| Accelerators | 144 accelerators | 2026-10 | Operator stated | AWS product page: up to 144 Trainium3 chips per Trn3 UltraServer; maximum configuration, undated page | aws.amazon.com |
The read
AWS's Trn3 UltraServer, which the product page says scales to up to 144 Trainium3 chips on NeuronLink-v4 (2 TB/s per chip) through a NeuronSwitch-v1 all-to-all fabric, with up to 362 MXFP8 PFLOPs (the page labels the same figure FP8 in one place), up to 20.7 TB of HBM3e and 706 TB/s of memory bandwidth; each chip has 144 GB of HBM3e at 4.9 TB/s. The page puts UltraServers in EC2 UltraClusters 3.0 of hundreds of thousands of chips but gives no date, region list or fleet size, and the status is as of the fetch date. Quantity is 'up to 144', the maximum configuration, not a count of deployed machines. Product configuration with no stated location, so the site is blank and USA is the operator's home country. Peak figures are low-precision vendor peaks, not HPL. Successor to aws-trn2-ultraserver, which holds 64 chips.
Timeline
No dated events recorded.
Change history
- 2026-10-17 Added AWS Trainium3 UltraServer, up to 144 chips (product configuration)
Source check
We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.
| Cited source | Result | Found on the page |
|---|---|---|
| aws.amazon.com | names system only | · counts: 144 |
Checked 2026-10-18 by pnpm verify. Re-run it and the table changes with the web.