Legal
Terms of Service
Inferfly · inferfly.ai
These Terms constitute a legally binding agreement between you and Inferfly governing your access to and use of the Inferfly platform, APIs, and related services.
Last updated: April 4, 2026 · Effective date: April 4, 2026
These Terms of Service ("Terms") constitute a legally binding agreement between you ("Customer", "you", or "your") and Inferfly ("Inferfly", "we", "us", or "our"), a sole proprietorship registered in India, governing your access to and use of the Inferfly platform, APIs, and related services (collectively, the "Service").
By accessing or using the Service, creating an account, or clicking "I Agree," you acknowledge that you have read, understood, and agree to be bound by these Terms. If you are using the Service on behalf of an organization, you represent that you have the authority to bind that organization to these Terms.
1. Definitions
"API" means the application programming interface(s) provided by Inferfly through which Customers access inference endpoints.
"API Key" means the unique authentication credential issued to a Customer for accessing the Service.
"Deployment" means a provisioned inference endpoint running a specific model configuration, accessible via a unique subdomain (e.g., {deployment-id}.api.inferfly.ai).
"GPU Hour" means one hour of access to a single GPU instance powering a Deployment. GPU Hours are the primary unit of billing and are consumed in whole-hour increments. Any partial use of a GPU Hour is billed as a full GPU Hour.
"Model" means an open-source large language model, embedding model, or reranking model deployed through the Service.
"Customer Data" means any data, including prompts, inputs, outputs, and fine-tuned model weights, that a Customer submits to or generates through the Service.
"Add-on Services" means optional supplementary services offered alongside inference endpoints, including but not limited to managed Open WebUI instances.
2. Description of Service
Inferfly provides managed GPU inference endpoints for open-source large language models. The Service includes:
- Inference Endpoints: OpenAI-compatible API endpoints for text generation, embedding, and reranking workloads.
- Managed Infrastructure: Provisioning, configuration, and monitoring of GPU-backed model deployments, including vLLM configuration, tensor parallelism, and operational management.
- Monitoring and Metrics: Access to deployment performance metrics, including throughput, latency, and GPU utilization.
- Add-on Services: Optional managed services such as Open WebUI chat interfaces, pre-configured to connect to Customer inference endpoints.
Inferfly abstracts infrastructure complexity so that Customers receive production-grade endpoints without requiring GPU operations expertise. The underlying GPU infrastructure is provisioned through third-party compute providers.
3. Account Registration
3.1 Eligibility
The Service is available only to individuals and entities located in the United States of America and the Republic of India. By accessing or using the Service, you represent and warrant that you are a resident of, or an entity organized and operating in, one of these countries. Inferfly reserves the right to restrict access from other jurisdictions without notice.
You must be at least 18 years of age and have the legal capacity to enter into a binding agreement to use the Service. If you are using the Service on behalf of an entity, you must have the authority to accept these Terms on that entity's behalf.
3.2 Account Responsibilities
You are responsible for:
- Providing accurate, current, and complete registration information.
- Maintaining the confidentiality of your API Keys and account credentials.
- All activities that occur under your account, whether or not authorized by you.
- Notifying Inferfly immediately at security@inferfly.ai if you become aware of any unauthorized use of your account or API Keys.
3.3 API Key Security
API Keys are hashed before storage and cannot be retrieved after initial issuance. If a key is compromised, you must regenerate it immediately through the platform console. Inferfly is not liable for any loss or damage arising from your failure to secure your API Keys.
4. Acceptable Use
4.1 Permitted Use
You may use the Service for lawful purposes in accordance with these Terms. This includes deploying open-source models for inference, integrating endpoints into your applications, and using Add-on Services as intended.
4.2 Prohibited Conduct
You agree not to:
- Use the Service to generate, distribute, or store content that is illegal, harmful, threatening, abusive, harassing, defamatory, or otherwise objectionable under applicable law.
- Use the Service to develop, train, or deploy models intended to produce child sexual abuse material, biological or chemical weapon instructions, or any content facilitating violence against individuals or groups.
- Attempt to gain unauthorized access to Inferfly's infrastructure, other Customers' deployments, or third-party systems through the Service.
- Reverse-engineer, decompile, or disassemble any component of the Service.
- Use the Service to send unsolicited communications (spam) or conduct phishing attacks.
- Resell or sublicense access to the Service without Inferfly's prior written consent.
- Circumvent or attempt to circumvent rate limits, usage quotas, or other technical restrictions.
- Use the Service in any manner that could damage, disable, overburden, or impair the Service or interfere with any other party's use.
- Deploy models in violation of their respective open-source licenses.
4.3 Model License Compliance
Customers are solely responsible for ensuring that their use of any model deployed through the Service complies with the applicable open-source license terms (e.g., Apache 2.0, Llama Community License, Gemma Terms of Use). Inferfly does not grant any license to models and makes no representations regarding the permissibility of specific use cases under any model license.
4.4 Enforcement
Inferfly reserves the right to suspend or terminate any Deployment or account that violates these Terms, with or without notice, depending on the severity of the violation.
5. Customer Data
5.1 Ownership
You retain all rights, title, and interest in your Customer Data. Inferfly does not claim ownership of any Customer Data.
5.2 Limited License
You grant Inferfly a limited, non-exclusive license to process Customer Data solely as necessary to provide and maintain the Service. This includes proxying requests to inference endpoints, logging metadata for billing and monitoring, and caching as necessary for performance optimization.
5.3 No Training on Customer Data
Inferfly does not use Customer Data to train, fine-tune, or improve any machine learning models, whether Inferfly's own or third-party models.
5.4 Data Handling
- Prompt and Response Data: Inference requests and responses are processed in real time and are not persistently stored by Inferfly beyond what is necessary for request proxying and metering. Request metadata (timestamps, token counts, status codes) may be retained for billing and monitoring purposes.
- Fine-Tuned Model Weights: If you deploy a fine-tuned model, the model weights reside on the provisioned GPU infrastructure for the duration of the Deployment. Upon termination of a Deployment, model weights on the infrastructure may be deleted in accordance with compute provider data retention policies.
- Logs: Request logs (excluding prompt and response content) may be retained for operational, debugging, and billing purposes for up to 90 days.
5.5 Data Residency
Inferfly may offer deployment options in specific geographic regions, including India-based GPU infrastructure, to assist Customers with data residency requirements. It is the Customer's responsibility to select the appropriate region for their compliance needs. Inferfly does not guarantee compliance with any specific data residency regulation solely by virtue of infrastructure location.
6. Payment and Billing
6.1 Pricing
Service pricing is based on GPU instance type, deployment duration, and any applicable Add-on Services. Current pricing is published on the Inferfly website and may be updated from time to time. Price changes will not apply retroactively to existing active Deployments but will apply to new Deployments and renewals.
6.2 GPU Hours
The primary unit of billing for the Service is the GPU Hour, defined as one hour of access to a single GPU instance powering a Deployment. Customers purchase GPU Hours in advance to provision and maintain Deployments.
6.3 Payment Processing
Payment processing is handled by our Merchant of Record. By making a payment, you also agree to the payment processor's terms of service. Inferfly supports payments via credit/debit cards and UPI (for Indian customers).
6.4 Billing and Consumption
GPU Hours are consumed for the entire duration a Deployment is active, regardless of whether inference requests are being made. A GPU Hour begins at the moment the Deployment is started or a new hour period commences and is consumed in full at that point. If a Deployment is active for any portion of a given hour, the entire GPU Hour is consumed and billed. For example, if a Deployment runs for 3 hours and 1 second, 4 GPU Hours are consumed. You are responsible for stopping Deployments when they are no longer needed to avoid continued consumption of GPU Hours. Billing details, including GPU Hours purchased, consumed, and remaining, are available through the platform console.
6.5 Taxes
All prices are exclusive of applicable taxes unless stated otherwise. The Merchant of Record will apply Goods and Services Tax (GST) for Indian transactions and any other applicable taxes as required by law.
6.6 Refunds
Refunds are available only for whole, unused GPU Hours. Any GPU Hour that has been partially consumed (i.e., a Deployment was active for any portion of that hour) is non-refundable. For example, if you purchased 100 GPU Hours and consumed 42 full hours plus 10 minutes of an additional hour, 43 GPU Hours are considered consumed, and a refund may only be issued for the remaining 57 unused GPU Hours. Refund disputes must be submitted to support@inferfly.ai. Unused whole hours will be available as credits to your account for future deplouyments. Refunds for Add-on Services are evaluated on a case-by-case basis.
6.7 Overdue Payments
If payment is overdue, Inferfly may suspend your access to the Service after providing reasonable notice. Continued non-payment may result in termination of your account and deletion of associated Deployments.
7. Service Availability
7.1 Uptime
Inferfly strives to maintain high availability for all Deployments but does not guarantee uninterrupted, error-free service. The Service depends on third-party GPU compute providers and network infrastructure that are outside Inferfly's direct control.
7.2 Scheduled Maintenance
Inferfly may perform scheduled maintenance that could temporarily affect Service availability. Where practical, we will provide advance notice of scheduled maintenance through the platform console or email.
7.3 No SLA (Current)
Inferfly does not currently offer a formal Service Level Agreement (SLA) with guaranteed uptime commitments or financial credits. This may change in the future, and any SLA will be published separately and incorporated by reference into these Terms.
8. Intellectual Property
8.1 Inferfly IP
The Service, including its architecture, APIs, routing layer, control plane, monitoring systems, documentation, and branding, is the intellectual property of Inferfly. Nothing in these Terms transfers any Inferfly IP to you.
8.2 Open-Source Models
Models deployed through the Service are licensed under their respective open-source licenses. Inferfly does not claim any intellectual property rights over open-source models.
8.3 Customer Applications
Inferfly claims no rights over any applications, products, or services that you build using the Service.
9. Third-Party Services
The Service relies on third-party infrastructure providers for GPU compute, content delivery, and related services. Inferfly is not responsible for the performance, availability, or policies of third-party providers. Your use of the Service is subject to the limitations inherent in the third-party infrastructure.
10. Limitation of Liability
10.1 Disclaimer of Warranties
THE SERVICE IS PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTIES OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.
INFERFLY DOES NOT WARRANT THAT THE SERVICE WILL BE UNINTERRUPTED, SECURE, OR ERROR-FREE, THAT DEFECTS WILL BE CORRECTED, OR THAT THE SERVICE OR THE INFRASTRUCTURE SUPPORTING IT ARE FREE OF VIRUSES OR OTHER HARMFUL COMPONENTS.
10.2 Limitation of Liability
TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT SHALL INFERFLY, ITS PARTNERS, EMPLOYEES, OR AGENTS BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, OR PUNITIVE DAMAGES, INCLUDING BUT NOT LIMITED TO LOSS OF PROFITS, DATA, USE, GOODWILL, OR OTHER INTANGIBLE LOSSES, ARISING OUT OF OR IN CONNECTION WITH YOUR USE OF OR INABILITY TO USE THE SERVICE.
INFERFLY'S TOTAL AGGREGATE LIABILITY FOR ALL CLAIMS ARISING OUT OF OR RELATED TO THESE TERMS OR THE SERVICE SHALL NOT EXCEED THE AMOUNT YOU PAID TO INFERFLY IN THE TWELVE (12) MONTHS PRECEDING THE EVENT GIVING RISE TO THE CLAIM.
10.3 Acknowledgment
You acknowledge that the limitations of liability in this section are a fundamental element of the agreement between you and Inferfly, and that Inferfly would not provide the Service without such limitations.
11. Indemnification
You agree to indemnify, defend, and hold harmless Inferfly and its partners, employees, and agents from and against any and all claims, damages, losses, liabilities, costs, and expenses (including reasonable legal fees) arising out of or related to:
- Your use of the Service.
- Your violation of these Terms.
- Your violation of any applicable law or regulation.
- Your violation of any third-party rights, including model license terms.
- Any Customer Data processed through the Service.
- Any application or service you build using the Service.
12. Termination
12.1 Termination by Customer
You may terminate your account at any time by stopping all Deployments and contacting support@inferfly.ai. You remain responsible for any charges incurred prior to termination.
12.2 Termination by Inferfly
Inferfly may suspend or terminate your account or any Deployment immediately if:
- You breach these Terms.
- Your use of the Service poses a security risk to the Service or other Customers.
- You fail to pay any amounts owed.
- Required by law or regulatory order.
For terminations not related to breach, Inferfly will make reasonable efforts to provide advance notice.
12.3 Effect of Termination
Upon termination:
- Your access to the Service will cease.
- All active Deployments will be shut down.
- Customer Data, including model weights stored on deployment infrastructure, may be deleted. It is your responsibility to export or back up any data before termination.
- Sections of these Terms that by their nature should survive termination (including Limitation of Liability, Indemnification, and Governing Law) will survive.
13. Privacy
Inferfly's collection and processing of personal data is governed by our Privacy Policy, available at inferfly.ai/privacy-policy. By using the Service, you agree to the collection and use of information as described in the Privacy Policy.
For Customers subject to the Digital Personal Data Protection Act, 2023 (DPDP Act) of India, Inferfly will process personal data in accordance with applicable obligations under the Act.
14. Modifications to Terms
Inferfly reserves the right to modify these Terms at any time. Material changes will be communicated via email to the address associated with your account or through a notice on the platform. Continued use of the Service after the effective date of any modifications constitutes your acceptance of the updated Terms. If you do not agree with the modified Terms, you must stop using the Service and terminate your account.
15. Governing Law and Dispute Resolution
15.1 Governing Law
These Terms are governed by and construed in accordance with the laws of India, without regard to conflict of law principles.
15.2 Dispute Resolution
Any dispute arising out of or in connection with these Terms shall first be attempted to be resolved through good-faith negotiation. If the dispute is not resolved within thirty (30) days, it shall be referred to and finally resolved by arbitration administered under the Arbitration and Conciliation Act, 1996 of India. The seat of arbitration shall be in India. The language of arbitration shall be English.
15.3 Jurisdiction
Subject to the arbitration clause above, the courts in India shall have exclusive jurisdiction over any disputes arising from these Terms.
16. General Provisions
16.1 Entire Agreement
These Terms, together with the Privacy Policy and any applicable order forms, constitute the entire agreement between you and Inferfly with respect to the Service.
16.2 Severability
If any provision of these Terms is found to be unenforceable, the remaining provisions will continue in full force and effect.
16.3 Waiver
Failure by Inferfly to enforce any right or provision of these Terms shall not constitute a waiver of that right or provision.
16.4 Assignment
You may not assign or transfer your rights under these Terms without Inferfly's prior written consent. Inferfly may assign its rights and obligations under these Terms without restriction.
16.5 Force Majeure
Inferfly shall not be liable for any delay or failure to perform resulting from causes beyond its reasonable control, including but not limited to acts of God, natural disasters, war, terrorism, pandemics, labor disputes, government actions, power failures, internet or telecommunications failures, or failures of third-party compute providers.
16.6 Notices
Notices to Inferfly should be sent to legal@inferfly.ai. Notices to you will be sent to the email address associated with your account.
17. Contact
If you have questions about these Terms, please contact us at:
Inferfly
Email: legal@inferfly.ai
Website: https://inferfly.ai
You may also contact us through our general inquiry form.
Have questions?
If you have any questions about our Terms of Service, please don't hesitate to reach out.
Contact Us