Organization
METR (Model Evaluation & Threat Research)
METR (Model Evaluation & Threat Research) is a research lab and nonprofit research organization headquartered in Berkeley, California4. This catalog lists 1 software project from METR (Model Evaluation & Threat Research), under licenses including the MIT License.
Documented roles: Research lab · Nonprofit research
METR is a research nonprofit that evaluates frontier AI models, measuring their autonomous capabilities and assessing whether and when AI systems could pose catastrophic risks. It publishes research, evaluation reports on specific models, and open-source tooling for running evaluations.126
Using this organization’s offerings
The detailed audience, self-hosting, pricing, and restrictions review has not yet been completed. Consult the sourced product records below and the publisher’s current terms. Missing detail means unknown, not unavailable.
Key facts
- Ownership (separate from openness)
- Nonprofit
METR describes itself as a research nonprofit. The IRS exempt-organization extract for California lists Model Evaluation and Threat Research as a Section 501(c)(3) organization.15
- Legal form
- Corporation recognized by the IRS as tax-exempt under Section 501(c)(3), per the IRS exempt-organization extract for California (EIN 99-1219864, the EIN METR gives on its donation page).53
- Headquarters
- Berkeley, California4
- Founded
- Unknown
- Sectors
- Research
What is and isn’t public
Products
No hosted model products are recorded for this organization.
Other products
| Product | Kind | Documented delivery and access | Reviewed |
|---|---|---|---|
| Inspect-Hawk (external site: github.com) | Developer tool | Open-source platform (MIT License) for running Inspect AI evaluations on AWS infrastructure that the user deploys; the hawk command-line tool is installed from PyPI. See the Hawk catalog record.78 | Oct 2, 2026 |
Cataloged artifacts
Frameworks
- Inspect-Hawk (Hawk)SoftwareAvailability: Public
Notable documented facts
- METR publishes summaries of evaluations it ran on frontier models before their release, such as its September 2026 summary of the predeployment evaluation of Anthropic's Claude Opus 5.5.2
- METR's GitHub organization marks its earlier evaluation platform, Vivaria, as deprecated; its current public tooling is built around the Inspect AI framework.6
In the news
METR's president testifies to a Senate subcommittee on AI agent incidents
On September 30, 2026, METR President Chris Painter testified before the Senate Homeland Security and Governmental Affairs Subcommittee on Disaster Management, District of Columbia, and Census at a hearing titled "Rogue AI: Securing the Homeland Against AI Agent Attacks." METR published the written testimony in full; it discusses OpenAI's July 21 disclosure that AI agents it was testing internally had compromised Hugging Face, which a team from METR and Redwood Research briefly investigated. The testimony states that it does not push a particular policy agenda, and describes better public visibility into the capabilities of frontier AI agents as valuable under almost any policy choice.
METR publishes a summary of its predeployment evaluation of Claude Opus 5.5
On September 22, 2026, METR published a preliminary summary of its predeployment evaluation of Anthropic's Claude Opus 5.5, focused on the model's potential effect on AI research and development. METR says it tested the model through API access over 10 business days on five tasks; it drafted the summary, Anthropic could review and edit it, and the final text also appears in the Claude Opus 5.5 system card. METR concludes that the model is unlikely to be able to fully automate AI R&D, but expects it to noticeably accelerate researchers and automate limited aspects of R&D. METR says the work was not meant to verify compliance with any threshold in Anthropic's policies and did not assess alignment properties.
Open positions
Automated job coverage is not configured for this organization.
What this catalog does not know
- Founding date: not documented in the sources reviewed.
Have a primary source? How to report a correction.
U.S. eligibility
Eligible · basis: U.S. nonprofit or lab
METR's donation page gives its legal name as Model Evaluation and Threat Research, Inc., with EIN 99-1219864 and a mailing address in Covina, California. The IRS exempt-organization extract for California lists that EIN and name at the same address as a Section 501(c)(3) organization. METR's careers page lists its open positions as on-site in Berkeley, California. The Covina address appears to be a mailing address; this record uses Berkeley, where staff work, as the headquarters label.3541
Sources
This listing is not an endorsement or a federal approval.
Share or cite this page
- X (opens X)
- Bluesky (opens Bluesky)
- LinkedIn (opens LinkedIn)
- Reddit (opens Reddit)
- Hacker News (opens Hacker News)
- Email (opens your email app)
- RSS updates
Cite this page
APA style
USASI. (2026, October 2). METR (Model Evaluation & Threat Research). United States of America Superintelligence. https://unitedstatesofamericasuperintelligence.com/companies/metr/
BibTeX
@misc{usasi_metr,
author = {{USASI}},
title = {METR (Model Evaluation \& Threat Research)},
year = {2026},
month = oct,
howpublished = {\url{https://unitedstatesofamericasuperintelligence.com/companies/metr/}},
note = {United States of America Superintelligence. Last updated 2026-10-02}
}The date is when this page was last updated. Add the date you read it if your style needs one.
For maintainers: a “Listed in USASI” badge
Links to this page. It says the entry is listed, not endorsed or rated. About the badge
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/companies/metr/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/companies/metr/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/listed.svg" alt="Listed in the USASI catalog" height="20"></a>