Skip to main content

FAQs

Welcome to the aiXplain Frequently Asked Questions (FAQ) page. Here, you'll find answers to the most common questions about aiXplain's platform, services, and features. Whether you're new to aiXplain or looking to expand your knowledge, this page will guide you through the essential aspects of building, deploying, and managing agentic solutions. From understanding available assets to exploring pricing models and team features, we've got you covered.

What can I do on aiXplain?​

On aiXplain, you can build and deploy agentic solutions, including autonomous agents, multi-agent systems, and pipelines(structured workflows). These solutions are powered by assets, which are the foundational building blocks available on aiXplain’s Discover marketplace. With aiXplain services, you can ground, customize, and evaluate your assets and solutions to create scalable, high-quality AI-powered systems.

  • Discover marketplace

    • Contains over 40,000 assets, including agents, pipelines, LLMs, AI models, utilities, and APIs.
    • Allows you to browse, try, and compare assets to find the best fit for your use case.

  • FineTune - Customize leading LLMs to match your specific requirements.

  • Benchmark: Evaluate LLMs and other AI models based on performance, cost, and efficiency.

  • Index: Build a vector database to enable retrieval-augmented generation (RAG), one of the most powerful architectures for modern AI applications.

All of these capabilities are accessible through the SDK, with some available as low/no-code tools in Studio. For further assistance, you can chat with Bel Esprit to help build solutions, select the right assets, or provide architecture recommendations to kickstart your project.

How do I start on aiXplain?​

  1. Register for an account—get free credits, no credit card required.
  2. Create a team and add team members (optional).
  3. Generate an API key to run aiXplain assets and services.
  4. Check out the Quickstart Guide in the documentation to start building agentic solutions.
  5. Find assets for your agent by visiting the aiXplain Marketplace (Discover).

Assets​

What types of assets are available on aiXplain?

aiXplain’s marketplace offers over 40K assets, including AI models, utilities, metrics, agents, pipelines, indexes, datasets and files. You can find a wide range of AI models such as large language models, image generation, speech recognition, speech synthesis, text analysis, object recognition, and many more. Utilities are ideal for connecting data sources and API services to your agents. Currently, aiXplain features over 50 vendors and hosts, offering both open and closed-source assets.

Can I onboard my own assets to aiXplain?

Yes, you can onboard your own models, utilities and data to aiXplain, either as private assets (soon as public assets with the opportunity to monetize). Additionally, if you notice an asset missing from our marketplace, we can onboard it on demand at no cost. ​If you want to onboard your LLM or any other genAI model, we can onboard it on aiXplain marketplace as long as it is hosted (e.g., on Fireworks, together.ai, AWS, Azure, GCP etc.)

Can I integrate or access any external vector databases with aiXplain?

aiXplain allows you to create and host a Qdrant vector database using its Index service: aiR. You can integrate an external vector database by calling it using our custom code as a tool feature in agents, or the script node in pipelines.

Pricing​

What is aiXplain’s pricing model?

Payments are made using aiXplain credits, where 1 credit = $1. We accept Google Pay and credit card payments via Stripe.

aiXplain’s pricing is usage-based, based on data volume, with asset costs matching vendor prices and no additional fees for direct asset calls. When assets are used in deployed solutions, such as agents or pipelines, a 20% service fee is added to the total cost. For businesses we provide subscription models and advanced features.

Can I transfer credits from my personal account to my team?

No, you cannot. During the hackathon, you will get a team voucher worth $10 in credits that you can add to your team’s wallet.

Multitenancy​

What features are available to teams on aiXplain?

Teams on aiXplain have access to dedicated storage (5 GB per team), can include an unlimited number of members, manage private assets, and create up to 10 active API keys. Each team has its own separate wallet for billing, and team assets are private by default, accessible only through the team's member API keys.

Does aiXplain have role-based permissions?

Yes, aiXplain supports role-based permissions for effective team collaboration. You can create teams on aiXplain. Teams can have different roles, including Member, Admin, and Owner, each with specific permissions. For example, Admins and Owners can manage API keys, set usage and rate limits, invite/remove members, and handle billing. Members can create and manage assets, bookmark assets, and view team activity. The platform allows teams to be structured based on these roles to ensure proper control and access to resources.​

Deployment​

Where can I deploy my assets (agents, pipelines, models, etc.)?

aiXplain assets run on the cloud. When you create a new asset, e.g., agent, pipeline, or fine-tune a model, it is automatically deployed to aiXplain’s cloud (AWS by default), and an inference API is generated for direct access.

For businesses, we offer enterprise-garde deployment options, including

  • Dedicated instances with secure networking (VPN) and fully managed services for dedicated resources with no shared compute, optimizing speed, availability, and uptime.
  • On-edge/on-premises deployment to a virtual private cloud or a local data center with private networking and controlled access environments for data residency and compliance.

Will aiXplain scale if my app explodes?

aiXplain is built on a scalable infrastructure and can handle millions of requests per minute with very high concurrency.

Integration and monitoring​

What integration methods does aiXplain offer?

aiXplain supports integration via API (CURL, Python, JavaScript), SDKs (Python, JavaScript, Swift), and OpenAI API for select models. You can also integrate any external service or data source as a tool inside agents and pipelines using simple Python code, making your solutions highly adaptable and extensible. You can also integrate aiXplain endpoints into automation tools like Zapier.

How can I monitor my aiXplain endpoints?

You can monitor your aiXplain endpoints through built-in telemetry stats available for every API call. These include success status, duration, cost, and data size for both input and output.

Team transactions provide an overview of all successful API calls, including key details such as data size, cost, asset ID, in a given timeframe. This allows teams to track usage, analyze spending and revenue.

Coming soon: monitor your agent using AgentOps.

Coming soon: view and download full API logs.

Security, privacy and compliance​

At aiXplain, trust is our foundation, ensuring flexible, secure, and compliant AI deployment with full data control. Built with industry-leading encryption, secure architecture, and compliance measures, our platform guarantees data privacy, no training on user data, and complete operational confidence for businesses and developers.

Data security and privacy

Your data is yours. aiXplain follows strict data security protocols to ensure confidentiality, integrity, and compliance with enterprise standards.

Your Data, Your Control.

aiXplain does not store, log, or use your prompts or responses for model training. Fine-tuned and customized models remain fully private and are never shared with third parties. You can opt-out of agent session storage or route the data directly to your own database.

Flexible and secure deployment

aiXplain provides deployment flexibility to meet your security, compliance, and performance needs:

Multi-tenant cloud (default)​

AI models and agents are securely deployed in aiXplain’s multi-tenant cloud on AWS, abstracting the complexity of managing compute resources, while maintaining security, scalability, and global availability.

Dedicated infrastructure​

Choose reserved compute instances for isolated, high-performance deployments.

Private cloud/on-premises deployment​

Deploy aiXplain's AI stack in your own private cloud or on-premises for full control over data residency and compliance.

Whether you need a fully managed AI service or a private AI environment, aiXplain adapts to your business needs.

Proximity deployment​

To meet regional compliance and performance needs, aiXplain partners with Azure, GCP, Oracle, NVIDIA, and others to enable proximity deployment, ensuring low latency, high performance, and data sovereignty. Businesses can deploy AI models closer to their operations, whether in KSA, UAE, or other regions, while maintaining security and regulatory compliance.

Secure AI architecture & connectivity

Secure by design​

All data is encrypted in transit (TLS 1.2 minimum) and at rest using industry-leading security standards to ensure confidentiality and integrity. Granular API monitoring provides full visibility into AI usage for both your organization and customers—without requiring an external cloud account.

Maximum performance meets security​

For customers needing high-performance, secure, and isolated deployments, aiXplain offers dedicated instances with provisioned capacity, ensuring predictable performance, low latency, and enhanced security. On demand, aiXplain provides dedicated VPNs for these instances or other infrastructure, enabling private, secure access without internet exposure.

Compliance and governance

aiXplain is designed for businesses that require strict compliance with industry standards, ensuring security, transparency, and governance in AI operations. The platform is SOC 2 Type I compliant (SOC 2 Type II certification by mid-2025). Audit and monitoring tools provide visibility and control over AI interactions, enabling organizations to maintain compliance and security. Additionally, compliance-aligned AI guardrails offer automated safeguards to meet regulatory requirements and prevent AI misuse, ensuring responsible and ethical AI deployment.

Where can I reach you?​

We’re available all the time on Discord, or on our support email - care@aixplain.com.