Terms of Service

Last updated: September 29, 2026

1. Acceptance of Terms

These Terms of Service (the "Terms") constitute an agreement between you and Unfork ("we") regarding access to and use of the unfork.ai website and its public Amos decision-model endpoint (the "Service"). By accessing, browsing, or using the Service in any way, you agree to be bound by these Terms. If you do not agree, please stop using the Service immediately.

2. Service Description

• The Service includes: website content (docs, benchmark reports, model cards, etc.), the public evaluation endpoint api.unfork.ai/v1/systemone, and related open-source repositories (weights, code, and evaluation data).
• The public endpoint is intended for evaluation and integration testing. We may adjust rate limits, change, or partially discontinue the Service at any time, with notice where reasonably practicable.
• Enterprise services such as private deployment and custom support are negotiated separately by email and governed by a separately signed agreement.

3. Acceptable Use

When using the Service, you agree not to:

• violate any applicable law, regulation, or third-party rights;
• stress-test, scrape, evade rate limits maliciously, or otherwise disrupt normal operation of the Service;
• use the Service to generate, distribute, or facilitate unlawful, infringing, fraudulent, or harmful content;
• reverse engineer or circumvent security and access controls of the Service (except as permitted by the applicable open-source license);
• resell, sublicense, or commercially provide the Service to third parties without our written permission.

4. Intellectual Property

• The website copy, design, graphics, and page structure are owned by us and protected by copyright law.
• Amos model weights, code, and evaluation data are provided under their respective open-source licenses (see LICENSE / model card in each repository); read and comply with those license terms before use.
• Text you submit while using the Service remains yours; you grant us the limited processing rights necessary to provide the Service, improve the model, and conduct public evaluations.

5. Disclaimer

The Service is provided "as is" without warranties of any kind, express or implied, including merchantability, fitness for a particular purpose, and non-infringement. Model outputs (including decision and calibrated_confidence) are for reference only and do not constitute legal, medical, financial, or other professional advice; you are responsible for business decisions based on them.

6. Limitation of Liability

To the maximum extent permitted by law, we are not liable for any indirect, incidental, special, punitive damages, or lost profits. Our aggregate liability arising from the Service shall not exceed the fees you actually paid for the relevant service in the 12 months preceding the claim (or zero if none).

7. Service Availability

We strive to keep the Service available but do not promise uninterrupted or error-free operation. The public endpoint may be temporarily unavailable (HTTP 503) due to maintenance or overload; for production-grade availability, consider private deployment with your own high-availability architecture.

8. Changes to Terms

We may update these Terms from time to time. Changes will be posted on this page with an updated "Last updated" date; continued use of the Service constitutes acceptance of the updated Terms. We will provide reasonable notice of material changes.

9. Contact

Questions about these Terms: hello@unfork.ai.

These Terms are governed by the law of our place of establishment. If any provision is held invalid, the remainder stays in effect. In case of discrepancy between the Chinese and English versions, the Chinese version prevails.