USA · operational · ai training

TX-GAIN

Also known as TX-Generative AI Next; LLSC TX-GAIN

Operated by MIT Lincoln Laboratory at MIT LLSC Holyoke .

Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1092 of 1153 systems have a readable citation that names them, and 40 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: 13.39 PFlop/s
Rpeak: 41.82 PFlop/s
Rmax ÷ Rpeak: 32.0%
Cores: 99,216
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	- HP DL385 cluster (TOP500)	Hewlett Packard Enterprise HPENYSE	-	Reported	top500.org
Accelerator	NVIDIA H100 SXM5 Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W NVIDIA H100 (TOP500); MIT cites more than 600 NVIDIA GPUs without the model	NVIDIA NVDANASDAQ	-	Reported	top500.org news.mit.edu
CPU	- AMD EPYC 9254 24C 3.9GHz	AMD AMDNASDAQ	-	Reported	top500.org
Operating system	Ubuntu Server Operating systems Linux distribution · Open source, free with paid support tiers Ubuntu	Canonical	-	Reported	top500.org

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	600 accelerators	2025-10	Operator stated	MIT release says more than 600 NVIDIA GPU accelerators; 600 is a lower bound, not an exact count.	news.mit.edu

The read

The MIT Lincoln Laboratory Supercomputing Center's generative-AI system at its Holyoke, Massachusetts data centre, successor to TX-GAIA. MIT announced it on 2 October 2025 as the most powerful AI supercomputer at any US university, with more than 600 NVIDIA GPUs and a peak of two AI exaflops, and says it came online in summer 2025. TOP500 lists an HPE DL385 cluster with AMD EPYC 9254 24-core 3.9 GHz hosts, NVIDIA H100 accelerators and 100G Ethernet under Ubuntu: 99,216 cores, Rmax 13.39 PFlop/s, Rpeak 41.82 PFlop/s. The MIT article does not name the GPU model or node counts, so no accelerator quantity is recorded. TOP500 ranks: 114th (June 2025), 131st (November 2025), 150th (June 2026).

Timeline

2025 First operational source
2025-06 List debut at #114 (June 2025) source

Public list appearances

Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.

#114: TOP500, June 2025 →

Site & facility

MIT LLSC Holyoke

Location: Holyoke, MA, USA

news.mit.edu developer.nvidia.com

Change history

2026-10-02 Added TX-GAIN, MIT Lincoln Laboratory's H100 AI supercomputer

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
news.mit.edu	names system only	-
top500.org	names system and part	NVIDIA H100 SXM5, Ubuntu Server

Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.