1. Definitions

1.1 "API" means the application programming interface through which you can access and use the AI Models.

1.2 "API Key" means the unique key issued to you to access the API services.

1.3 "Input" means any data or content, such as text, images or audio, that you submit to an AI Model through the API. Input is preprocessed into a format that the AI Model can process before it is transmitted.

1.4 "Output" means any content generated by an AI Model in response to your Input. Output may take various forms, such as text, embedding vectors or analysis results, depending on the type of service requested.

1.5 "User" or "you" means a person who has registered as a member under the Showroom Terms, has agreed to these Terms and uses the Platform.

1.6 "AI Model" means an artificial intelligence model that we make available through the Platform.

2. Intellectual Property

2.1 Subject to these Terms, we grant you a non-exclusive, non-transferable, revocable right to use the Platform for non-commercial purposes. You may not use any AI Model or Output for commercial purposes, or enable any third party to call an AI Model or submit Input to an AI Model, including by integrating the API into your products or services. You are solely responsible for any breach of this Section 2.1 and its consequences.

2.2 To the extent permitted by applicable law, you retain your rights in your Input and hold the rights in the Output that you generate in the course of your use of the Platform.

2.3 You grant us a perpetual, worldwide, non-exclusive, royalty-free right to use your Input and Output. We may use your Input and Output to train AI Models, and may reproduce, process, analyze, edit, combine, distribute, modify and adapt them (including by creating derivative works), and you agree not to assert any moral rights or similar rights against us in connection with that use. However, if you opt out through the Platform of having your Input and Output used to train AI Models, we will not use them for that purpose after you opt out. Any portion of your Input and Output that was incorporated into an AI Model through training before you opted out cannot technically be separated out and deleted.

2.4 You represent and warrant that you have lawfully obtained all rights, permissions and consents necessary to submit your Input and for us to process and use it in accordance with these Terms.

2.5 You may not sublicense, sell, lease, rent, distribute or otherwise transfer the Platform's API services or any AI Model to any third party, or resell or otherwise provide them to any third party in a way that allows the third party to call or use an AI Model directly.

2.6 These Terms do not transfer to you any ownership rights in the AI Models or in any of our trademarks, service marks or logos. We do not provide the weights, parameters, source code or other components of the AI Models, and all rights not expressly granted in these Terms are reserved by us. Even if the weights of an AI Model have been released under a separate license, these Terms govern your use of that AI Model through the Platform.

2.7 If you provide us with feedback on the Platform or any AI Model, you agree that we may use and exploit that feedback without restriction or compensation to you.

3. API Keys

3.1 If your agreement with us terminates, we may deactivate your API Key.

3.2 If you suspect that your API Key has been leaked, you must notify us immediately and request a new API Key. If there are signs of a possible leak, such as access from unusual locations, unusual usage patterns or a sudden increase in usage, we may suspend the API Key without prior notice as a precaution.

3.3 You must not share, transfer or lend your API Key to any third party, and any use of your API Key will be treated as your own use.

3.4 API Keys are valid for 30 days from issuance and are automatically deactivated when they expire.

3.5 If you do not log in to the Platform for 90 days, we may mark your account as dormant and deactivate your API Key. You can reactivate a dormant account by logging in again.

4. Restrictions

4.1 Requests may be blocked automatically if abnormal traffic patterns are detected, such as repeated identical requests, random parameter testing or sudden spikes in requests per second.

4.2 You may not use the API to:

(a) request information on creating malware or viruses, including the generation or explanation of code for system intrusion, data theft, denial-of-service attacks or similar activities;

(b) commit identity theft or fraud, including by seeking financial gain through impersonating others or generating false information;

(c) reproduce or imitate copyrighted content without permission;

(d) facilitate illegal transactions or services, including by generating content that supports transactions in controlled substances, weapons or illegal services;

(e) circumvent an organization's internal security policies, including by seeking ways to bypass firewalls, DLP systems or similar safeguards;

(f) request content that promotes discrimination, hatred or violence, or that encourages discriminatory speech or violent acts against particular groups;

(g) develop, train or improve, directly or through a third party, an AI model that competes with our AI Models using Output or any other part of the Platform, or provide Output to a third party to enable it to do so;

(h) attempt to extract or reproduce the weights, architecture or training data of an AI Model through automated high-volume calls or other means, reverse engineer an AI Model, or bypass or circumvent technical protection measures applied to an AI Model;

(i) circumvent usage limits by using multiple accounts or API Keys or by other means;

(j) exploit vulnerabilities in the Platform, or conduct load testing or security testing without our approval;

(k) generate, disseminate or amplify false or misleading information, such as fake news, disinformation or inflammatory information;

(l) attempt to circumvent or disable safety measures, such as content filters, applied to an AI Model or the Platform; or

(m) do anything else that violates applicable laws, the Showroom Terms or these Terms.

4.3 To provide a stable and secure service, we may apply safety measures such as content filters, and may monitor your use of the Platform to the extent necessary to determine whether you are complying with these Terms.

5. Disclaimer of Warranties

5.1 THE AI MODELS ARE PROVIDED "AS IS." IN PARTICULAR, WE DO NOT WARRANT THE FOLLOWING:

(a) ACCURACY: THE ACCURACY OF INFORMATION IN OUTPUT DEPENDS ON THE TRAINING DATA AND THE PERFORMANCE OF THE AI MODEL, AND OUTPUT MAY CONTAIN INCORRECT INFORMATION (HALLUCINATIONS).

(b) TIMELINESS: OUTPUT MAY NOT REFLECT INFORMATION THAT BECAME AVAILABLE AFTER THE AI MODEL WAS TRAINED, AND INFORMATION ON RECENT ISSUES OR EVENTS MAY BE INACCURATE.

(c) COMPLETENESS: OUTPUT MAY NOT INCLUDE ALL INFORMATION RELEVANT TO YOUR QUERY, AND ONLY PARTIAL ANSWERS MAY BE PROVIDED TO COMPLEX QUESTIONS.

(d) CONSISTENCY: THE SAME QUESTION MAY NOT PRODUCE EXACTLY THE SAME RESPONSE EACH TIME, AND RESULTS MAY VARY DEPENDING ON THE CONTEXT OR HOW THE INPUT IS COMPOSED.

5.2 IF YOU USE AN AI MODEL TO MAKE IMPORTANT DECISIONS, YOU SHOULD VERIFY THE OUTPUT AGAINST OTHER SOURCES, AND WE ARE NOT RESPONSIBLE FOR ANY ISSUES ARISING FROM SUCH USE.

6. Information We Collect

6.1 We automatically collect the following information generated in the course of your use of the Platform:

(a) client request information: API request identifier, IP address, requested API URL, HTTP headers, API Key and Input; and

(b) server response information: HTTP response code, server processing result code, response data size (bytes), response token count, server processing time and Output.

6.2 We use this information for error diagnosis and usage pattern analysis, including to identify performance bottlenecks, adjust resources based on usage, detect common error patterns and detect abnormal usage.

7. Your Responsibilities

7.1 You are solely and ultimately responsible for any use of Output generated through the Platform, including any application, disclosure, distribution, transmission, reproduction or modification of Output and any derivative works you create from it.

7.2 You must verify for yourself that Output complies with all applicable laws, regulations and guidelines in Korea and elsewhere, including intellectual property laws (such as copyright, patent and trademark laws), the Personal Information Protection Act, the Act on Promotion of Information and Communications Network Utilization and Information Protection, the Criminal Act, unfair trade laws and the export control laws of each country. You are solely responsible for any civil, criminal, administrative or other legal liability arising from any failure to comply or to verify compliance.

7.3 Before you make Output public or provide it to a third party, you must review and edit it to confirm, among other things, that it: (i) is factually accurate; (ii) is legally sound and complies with applicable regulations; (iii) does not infringe the rights of others; (iv) does not contain socially or ethically inappropriate content; (v) does not risk promoting discrimination, hatred, bias or violence; (vi) is not defamatory and does not spread false information; and (vii) is not subject to any legal restriction on its disclosure or distribution.

7.4 Output may be subject to third-party rights (including, without limitation, copyrights, trademark rights and rights in a person's name or likeness, including rights of publicity). Where necessary, you must obtain valid permission or a license from the relevant right holder in advance.

7.5 If you use Output in a high-risk field, such as medicine, law, finance, construction, safety or politics, it must be reviewed and supervised by appropriately qualified professionals, and it must not be used for automated decision-making without human judgment and accountability.

7.6 We will not intervene in any objection, dispute, complaint, request for investigation, lawsuit or other issue raised by a third party regarding any content, data, product, service or other material created based on Output, and you agree not to seek any damages or costs from us, or to hold us liable, in connection with any such matter. If we suffer any loss or have to respond to a third-party claim as a result of your actions, you must immediately compensate us for all resulting costs (including litigation costs and attorneys' fees) and losses. The foregoing does not apply where the loss or claim is caused by our willful misconduct or gross negligence.

7.7 Without our prior written consent, you may not cause Output to cite, or be presented as associated with, the names, logos, service names, model names or similar identifiers of us or our affiliates, or use any expression that may lead others to believe that we are the source of Output or that Output represents our official position.

7.8 You must not include any other person's personal information in Input. If you do, you are responsible for the consequences, and if we suffer any loss or have to respond to a third-party claim as a result, you must immediately compensate us for all resulting costs (including litigation costs and attorneys' fees) and losses. The foregoing does not apply where the loss or claim is caused by our willful misconduct or gross negligence.

7.9 When using the Platform and Output, you must take all measures required under applicable laws, including the Framework Act on the Development of Artificial Intelligence and the Creation of a Foundation for Trust.

8. Eligibility

8.1 You must be 19 years of age or older to use the Platform.

8.2 You must accurately enter the country in which you are located when you sign up and promptly update it if it changes.

8.3 You must comply with the export control and economic sanctions laws of the Republic of Korea, the United States and other relevant countries, and you represent and warrant that neither you nor your organization is located in a sanctioned country or region or is included on any list of sanctioned persons under applicable laws.

9. Changes to the Platform

9.1 We may change the version of an AI Model or discontinue any AI Model at any time.

9.2 If we begin charging for the Platform, we will post the fees and any changes to them on our website.

10. Usage Limits

10.1 To provide a stable service, we may set usage limits, such as on the number of API calls and the number of tokens processed.

10.2 You can check your remaining usage through API response headers or other means.

10.3 We may provide additional benefits, such as higher usage limits, to certain Users.

11. Account Suspension

11.1 If you are suspected of violating the Showroom Terms or these Terms, we may, upon prior notice to you, take measures such as deactivating your API Key or suspending your account.

11.2 If you violate the Showroom Terms or these Terms, we may permanently suspend your account.

12. Miscellaneous

12.1 Sections 2.1 through 2.5, 2.7, 4.2(g) and (h), 5 and 7 survive the termination of your agreement with us.

12.2 These Terms are provided in Korean and English. If there is any discrepancy between the Korean and English versions, the Korean version prevails.

Effective Date

These Terms take effect on November 9, 2026.