FAQs

Getting Started

What can I do on aiXplain?

On aiXplain, you can build and deploy autonomous AI agents powered by the world's largest AI asset marketplace. Whether you need a single intelligent agent or a coordinated team of specialized agents, aiXplain provides everything you need to create production-ready AI solutions.

What you can build:

• Autonomous AI agents - Build intelligent agents that can reason, plan, and use tools to accomplish complex tasks
• Multi-agent systems - Create teams of specialized agents that work together to solve sophisticated problems
• Custom tools - Onboard your own tools, APIs, and integrations to extend agent capabilities
• Knowledge bases - Ingest your unstructured data (documents, PDFs, images, text) using AIR (AI Information Retrieval) to ground your agents in your specific context

What you can manage:

• Govern your assets - Control access, set policies, and enforce compliance across all your AI assets
• Monitor your assets - Track performance, usage, costs, and quality with comprehensive analytics and telemetry

How to get started:

• Studio - Visual interface for building agents with low-code/no-code tools
  • Browse - Explore over 40,000 models, tools, and integrations from 50+ vendors
  • Build - Create single agents or multi-agent systems
• SDK - Build programmatically with Python, JavaScript, or Swift

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 call aiXplain assets and build agents programmatically using the aiXplain SDK
3. Explore Studio to discover assets and build your first agent---no-code
4. Check out the Getting Started guide in the documentation

What is Studio?

Studio is aiXplain's visual interface for discovering assets and building AI agents. It has two main experiences:

• Browse - Explore and test models, agents, tools, and integrations
• Build - Create single agents or multi-agent systems (team agents)

Studio provides both no-code tools (visual interface) and full API integration for programmatic access.

What is Console?

Console is aiXplain's administrative interface for managing your account, teams, billing, and monitoring. Use Console to:

• Manage team members and roles
• Purchase and track credits
• View transaction history and analytics
• Generate and manage API keys
• Configure account settings

Access Console: console.aixplain.com

---

Assets

What types of assets are available on aiXplain?

aiXplain's marketplace offers over 40,000 assets across four main categories:

Models - AI models including:

• Large language models (LLMs)
• Text-to-speech and speech-to-text
• Image generation and computer vision
• Sentiment analysis and text classification
• Object detection and recognition
• Translation and multilingual models

Agents - Autonomous AI agents that can use tools and models to accomplish tasks. Can be single agents or multi-agent systems (team agents).

Tools - Capabilities that extend agent functionality:

• APIs and web services
• Data processing utilities
• Search and retrieval tools
• Database connectors

Integrations - Connectors that generate tools by linking external services:

• SQLite databases
• Python Sandbox for custom scripts
• MCP (Model Context Protocol) servers
• AIR Vector Database for RAG

aiXplain features over 50 vendors and hosts, offering both open-source and closed-source assets.

Can I onboard my own tools to aiXplain?

Yes, you can onboard your own assets to aiXplain as private assets via Integrations:

• Custom tools - Onboard your own APIs and services as tools via integrations:

  • Python Sandbox - Write custom Python scripts to connect to any API or service
  • MCP Server - Connect Model Context Protocol servers with your custom capabilities

• Your data

  • SQLite - Upload your databases for natural language querying
  • Vector database - Upload your documents, files, and unstructured data to AIR vector databases

Visibility: All assets you create are private by default, visible only to your team. Support for public assets with monetization opportunities is coming soon.

Need help onboarding? If you notice an asset missing from the marketplace or need assistance onboarding your own assets, contact aiXplain support at care@aixplain.com. We can onboard assets on demand at no cost.

What are public vs. private assets?

Public assets:

• Visible to all aiXplain users
• Available in the marketplace
• Cannot be edited by you
• May have usage costs set by vendors

Private assets:

• Visible only to your team
• Fully editable (name, description, configuration)
• Can be transferred between teams
• Can be deleted
• You control pricing and access

When you create agents or tools, they are private by default.

Can I integrate external vector databases with aiXplain?

Yes, you have two options:

Option 1: Use aiR Vector DB Integration

• Create a Qdrant vector database using aiXplain's Index service (AIR)
• Automatically hosted on aiXplain
• Integrated as a tool in agents

Option 2: Connect external vector databases

• Use the Python Sandbox Integration to connect to external databases
• Supports any vector database with an API
• Full control over your existing infrastructure

Can I integrate external SQL databases with aiXplain?

Yes, you have two options:

Option 1: Upload SQLite files

• Upload .sqlite or .db files directly to aiXplain
• Use the SQLite integration to generate a tool
• Agents can query the database using natural language

Option 2: Connect external databases

• Connect to PostgreSQL, MySQL, or other SQL databases
• Use a custom tool in agents with connection credentials
• Use Python Sandbox integration to write custom queries

---

Pricing and Credits

What is aiXplain's pricing model?

aiXplain uses a consumption-based pricing model where you only pay for what you use.

Credit system:

• 1 credit = $1 USD
• All usage measured in credits
• Transparent, predictable pricing

Asset pricing:

• Models: Priced by vendors (per token, per request, etc.) - no aiXplain markup
• Tools: Priced by vendors - no aiXplain markup
• Agents: Sum of all model and tool calls + 20% markup for orchestration

Example agent cost:

LLM call: 0.5 credits
Web search tool: 0.1 credits
Database tool: 0.2 credits
Subtotal: 0.8 credits
Markup (20%): 0.16 credits
Total: 0.96 credits

Payment methods:

• All major credit cards (Visa, Mastercard, Amex, Discover)
• Apple Pay
• Voucher codes

Plans:

• Team Plan - For collaborative teams
• Enterprise Plan - For large-scale deployments with custom requirements

Visit aixplain.com/pricing for detailed pricing information.

How do I add credits?

Method 1: Purchase with payment method

1. Go to Console → Credits and Billing
2. Click "Add Credits"
3. Enter amount to purchase
4. Select or add payment method
5. Complete transaction

Method 2: Redeem voucher code

1. Go to Console → Credits and Billing
2. Click "Redeem Voucher"
3. Enter voucher code
4. Credits added immediately

Can I transfer credits between workspaces?

No, credits cannot be transferred between workspaces (personal or teams) by users. Each workspace has its own separate credit balance.

If you need to transfer credits, contact aiXplain Support at care@aixplain.com for assistance.

Where can I view my transaction history?

You can view transactions in multiple locations:

Console → Transactions

• Comprehensive transaction history
• Detailed breakdown by asset, date, cost
• Filter and export capabilities
• Running balance

Studio → Validation Tab

• Real-time cost display during agent testing
• Breakdown by component (each model call, tool call)
• Detailed execution traces

Programmatically via API

• Transaction details included in API responses
• Track costs in your applications
• Build custom dashboards

What happens if I run out of credits?

(Information to be confirmed - please contact support for current policy)

Admins and Owners receive email alerts when credits are running low to help prevent service interruptions.

---

Teams and Access Control

What features are available to teams on aiXplain?

Teams provide collaborative workspaces with:

Asset management:

• Unlimited private agents, tools, and custom integrations
• Share assets across team members
• Transfer assets between teams

Member management:

• Unlimited team members
• Role-based access control (Owner, Admin, Member)
• Invite and remove members

Billing and credits:

• Separate credit balance per team
• Dedicated billing and transaction tracking
• Credit allocation and monitoring

API access:

• Up to 10 API keys per team
• Rate limiting controls (Admin feature)
• Usage tracking per key

Storage:

• 5 GB dedicated storage per team
• File uploads for agents and tools

Monitoring:

• Full transaction details (data volume, API calls, costs)
• Analytics dashboard
• Usage and performance metrics

Does aiXplain have role-based permissions?

Yes, aiXplain provides role-based access control (RBAC) with three roles:

Permission	Member	Admin	Owner
Asset Management
Create agents and tools	✅	✅	✅
Use and consume assets	✅	✅	✅
Access team's private assets	✅	✅	✅
Transfer assets	❌	✅	✅
Delete private assets	❌	✅	✅
Team Management
Invite members	❌	✅	✅
Remove members	❌	✅	✅
Assign roles	❌	✅	✅
Delete team	❌	❌	✅*
Rename team	❌	❌	✅*
API & Integration
Create/delete API keys	❌	✅	✅
Set rate limits	❌	✅	✅
Billing & Credits
Manage payment methods	❌	✅	✅
Purchase credits	❌	✅	✅
View transactions	❌	✅	✅
Receive billing alerts	❌	✅	✅

*Coming soon

How do I create a team?

1. Click the Spinner (colored circle) in the top-right corner
2. Click Manage Team
3. Click Create Team
4. Enter a unique team name
5. Choose whether to switch to the new team
6. Invite members and assign roles

Your new team starts with its own workspace, separate credit balance, and no assets.

Can I belong to multiple teams?

Yes, you can belong to unlimited teams. Each team has its own:

• Private assets
• Credit balance
• API keys
• Transaction history

Switch between teams using the Spinner menu in the top-right corner.

What's the difference between personal workspace and team workspace?

Feature	Personal Workspace	Team Workspace
Owner	You	Team Owner
Members	Only you	Multiple members with roles
Private Assets	Visible only to you	Visible to all team members
Credits	Your personal balance	Team's shared balance
API Keys	Your keys (up to 10)	Team's keys (up to 10)
Billing	You manage	Admins/Owners manage
Collaboration	Solo work	Team collaboration

---

Deployment and Infrastructure

Where can I deploy my assets?

aiXplain assets (agents, models, tools) run on the cloud. When you create an asset, it is automatically deployed and an inference API is generated for immediate access.

Default deployment:

• aiXplain's multi-tenant cloud (AWS by default)
• Automatic scaling and management
• Global availability

Enterprise deployment options:

Dedicated instances:

• Private networking
• Managed services
• Optimal performance and security
• Reserved compute resources

On-premise/Private cloud:

• Full infrastructure control
• Complete data residency
• Custom compliance requirements
• Private networking

Proximity deployment:

• Deploy closer to your operations (KSA, UAE, other regions)
• Partners: Azure, GCP, Oracle, NVIDIA
• Low latency and high performance
• Regional data sovereignty

Contact sales for enterprise deployment options.

Will aiXplain scale if my application has high traffic?

Yes, aiXplain is built on scalable infrastructure and can handle:

• Millions of requests per minute
• High concurrency across multiple endpoints
• Automatic scaling based on demand
• Global distribution for low latency

For enterprise applications requiring guaranteed performance, dedicated instances with provisioned capacity are available.

Can I use aiXplain in production?

Yes, aiXplain is designed for production use with:

• Enterprise-grade infrastructure
• High availability and uptime
• Automatic scaling
• Monitoring and analytics
• API keys with rate limiting
• SOC 2 Type I and II compliance

Thousands of production applications run on aiXplain.

---

Integration and APIs

What integration methods does aiXplain offer?

SDK integration:

• Python SDK (recommended)
• JavaScript SDK
• Swift SDK

REST API:

• cURL
• Standard HTTP/REST protocols
• OpenAI-compatible API (select models)

Direct integration:

• Webhook endpoints
• Automation tools (Zapier, Make, etc.)
• Custom integrations via Python Sandbox

Example - Python SDK:

import aixplain
aixplain.api_key = "your-api-key"

agent = aixplain.Agent("agent-asset-id")
response = agent.run("What is the weather today?")

Example - REST API:

curl -X POST https://api.aixplain.com/v1/agents/run \
  -H "Authorization: Bearer your-api-key" \
  -d '{"agent_id": "agent-asset-id", "query": "Hello"}'

How can I monitor my aiXplain endpoints?

Built-in telemetry for every API call:

• Success status
• Duration (latency)
• Cost (credits consumed)
• Data size (input and output)
• Timestamps (start and end)

Team transactions dashboard:

• All successful API calls
• Filter by date range, asset type
• Data volume and costs
• Asset ID and details
• Export capabilities

Agent validation traces:

• Real-time execution monitoring
• Component-level breakdown
• Input/output for each step
• LLM reasoning and thoughts
• Credit consumption per component

Programmatic monitoring:

• API responses include telemetry data
• Build custom dashboards
• Set up alerts and notifications
• Track usage programmatically

Access monitoring via:

• Console → Transactions
• Console → Analytics
• Studio → Validation tab
• API response metadata

Can I integrate external tools and services?

Yes, you can integrate any external service or API into your agents using:

Python Sandbox integration:

• Write custom Python scripts
• Call external APIs
• Process data with custom logic
• Full Python ecosystem available

Custom tools:

• Create tools from any API
• Connect to databases
• Integrate with internal systems

MCP Server integration:

• Connect Model Context Protocol servers
• Access standardized tool interfaces

This makes your agents highly adaptable and extensible.

---

Security, Privacy, and Compliance

How does aiXplain protect my data?

Data security:

• All data encrypted in transit (TLS 1.2 minimum)
• All data encrypted at rest
• Industry-leading security standards
• SOC 2 Type I and II compliant

Data privacy:

• aiXplain does not store, log, or use your prompts for training
• Fine-tuned models remain fully private
• No data sharing with third parties
• You can opt-out of agent session storage

Your data, your control:

• You own all your data
• Route data directly to your own database
• Complete control over data residency

Is aiXplain compliant with industry standards?

Yes, aiXplain is designed for businesses requiring strict compliance:

Certifications:

• SOC 2 Type I and II compliant
• Regular security audits
• Compliance-aligned AI guardrails

Governance features:

• Audit and monitoring tools
• Full visibility into AI interactions
• Automated safeguards for regulatory requirements
• Prevent AI misuse

Enterprise compliance:

• Dedicated instances for isolated deployments
• Private cloud/on-premise options
• Regional data residency (proximity deployment)
• Custom compliance configurations

What deployment options meet security requirements?

Multi-tenant cloud (default):

• Secure deployment on AWS
• Abstracted compute management
• Security, scalability, global availability

Dedicated infrastructure:

• Reserved compute instances
• Isolated, high-performance deployments
• Predictable performance and low latency
• Enhanced security

Private cloud/On-premises:

• Full control over data residency
• Custom compliance requirements
• Private networking
• Complete infrastructure control

Secure connectivity:

• Dedicated VPNs available on demand
• Private access without internet exposure
• Granular API monitoring
• No external cloud account required

Can I use aiXplain in regulated industries?

Yes, aiXplain is suitable for regulated industries including:

• Healthcare - HIPAA considerations with private deployment
• Finance - SOC 2 compliant, dedicated instances available
• Government - On-premise and proximity deployment options
• Legal - Data residency and privacy controls

For specific compliance requirements, contact sales for custom configurations.

---

Support and Resources

Where can I get help?

Documentation:

• docs.aixplain.com - Comprehensive guides and references
• Getting started tutorials
• API reference documentation

Community:

• Discord - Active community, live support
• Ask questions, share solutions, get help

Direct support:

• Email: care@aixplain.com
• Available all the time
• Technical and billing support

In-platform:

• Feedback button (throughout Studio)
• Bel Esprit AI assistant
• Report issue on asset cards

How do I report bugs or request features?

Report bugs:

• Use the Feedback button in Studio or Console
• Email care@aixplain.com with details
• Report on Discord

Request features:

• Share suggestions on Discord
• Email care@aixplain.com
• Use feedback button with "[Feature Request]" in subject

Vote on features:

• Join Discord to participate in roadmap discussions
• Community-driven feature prioritization

Where can I learn more about using aiXplain?

Getting started:

• Getting Started Guide
• Studio walkthrough tutorials
• Sample agents and templates

Documentation:

• Browse documentation - Discover and test assets
• API reference - Technical documentation

Community resources:

• Discord community examples
• Blog posts and case studies
• Video tutorials (coming soon)

Professional services:

• Enterprise onboarding
• Architecture consultation
• Custom training

Contact sales for professional services: care@aixplain.com

---

Troubleshooting

My API key isn't working. What should I do?

Common causes:

• Key was deleted or expired
• Key copied incorrectly
• Using key from wrong workspace
• Insufficient credits in workspace

Solutions:

1. Verify key is active in Console → Settings → API Keys
2. Check expiration date
3. Ensure you copied the entire key (no spaces)
4. Confirm you're using correct workspace's key
5. Check credit balance in Console → Credits and Billing

If issue persists, delete the key and create a new one.

My agent is consuming too many credits. How can I optimize?

Optimization strategies:

1. Reduce tool count:

• Stay within 10-15 tools per agent
• Remove unnecessary tools
• Combine similar functionality

2. Select specific actions:

• Don't enable all tool actions
• Choose only required actions
• Fewer actions = faster, cheaper

3. Optimize instructions:

• Be specific about when to use tools
• Prevent unnecessary tool calls
• Use validation to test behavior

4. Choose efficient models:

• Balance cost vs. performance
• Smaller models for simple tasks
• Use streaming when appropriate

5. Set iteration limits:

• Configure maximum iterations
• Prevent runaway loops

6. Monitor in validation:

• Track cost per component
• Identify expensive operations
• Test before deploying

I can't find an asset I need. What should I do?

Options:

1. Search thoroughly:

• Use search bar with different keywords
• Browse by category
• Filter by vendor

2. Request onboarding:

• Email care@aixplain.com
• Specify the asset you need
• aiXplain can onboard it at no cost

3. Create custom solution:

• Use Python Sandbox for custom scripts
• Connect external APIs as tools
• Build custom integrations

4. Check documentation:

• Asset may be available under different name
• Review vendor documentation

How do I get enterprise features?

Enterprise features include:

• Dedicated instances
• On-premise/private cloud deployment
• Custom compliance configurations
• Higher rate limits
• Dedicated support
• Custom contracts and SLAs

Contact sales:

• Email: care@aixplain.com
• Subject: "Enterprise Inquiry"
• Include: use case, scale, requirements

Sales team will provide:

• Custom pricing
• Architecture recommendations
• Compliance guidance
• Implementation support

---

Have more questions? Join our Discord or email care@aixplain.com