AI Models

Choose the right AI model for your needs. CKEditor AI provides access to leading AI models, each optimized for different use cases and performance requirements.

CKEditor AI provides agent models that automatically select the optimal AI model for your specific task, balancing speed, quality, and cost. Agent models are internally registered and versioned, with the exact model name varying by compatibility version, for example, agent-1.

Benefits:

• Automatic optimization for each request type.
• No need to understand individual model capabilities.
• Cost-effective routing to appropriate models.
• Continuous improvement as new models become available.

For most use cases, we recommend using the agent model appropriate for your compatibility version and letting the system handle model selection.

All models support conversations, reviews, and actions. Each model has specific limits for prompt length, conversation length, file handling, and other constraints. For current specifications and availability in your environment, query the /v1/models endpoint.

The platform supports models from major AI providers including OpenAI, Google, and Anthropic. Each model has different capabilities, performance characteristics, and limitations. The system automatically handles model selection, failover, and optimization to provide the best experience for your use case.

Model Compatibility Versioning identifies sets of LLMs compatible with specific service versions. This ensures your application continues to work reliably as new models are added or existing models are updated. Each service version has a specific set of compatible models, with older versions remaining available for existing integrations and new models added to newer service versions, providing clear upgrade paths.

1. Versioned Model Sets – Each compatibility version represents a curated list of models available at a specific API version.
2. Stable Integrations – Your application continues working even as new models are added to the platform.
3. Controlled Upgrades – Upgrade to newer compatibility versions when ready to test and support new models.
4. Backward Compatibility – Older versions remain supported, allowing you to upgrade on your schedule.

You can check which models are available for your service version:

GET /v1/models/1

Currently, there is only one compatibility version available: 1. This version includes all currently supported models. As the platform evolves, new compatibility versions will be introduced to provide curated model sets for different API versions.

Example response:

{
  "items": [
    {
      "id": "gpt-4o",
      "name": "GPT-4o",
      "provider": "OpenAI",
      "allowed": false,
      "capabilities": {
        "webSearch": {
          "enabled": false
        },
        "reasoning": {
          "enabled": false
        }
      }
    },
    {
      "id": "gemini-2-5-pro",
      "name": "Gemini 2.5 Pro",
      "provider": "Google",
      "allowed": true,
      "capabilities": {
        "webSearch": {
          "enabled": true,
          "allowed": true
        },
        "reasoning": {
          "enabled": true,
          "allowed": false
        }
      }
    }
  ]
}

All models support conversations, reviews, and actions. Some models also support web search and reasoning capabilities.

Real-time web search with source attribution for current events, fact verification, and research. May increase response time.

When enabled, models can search the web before generating responses to access up-to-date information beyond their training data. You can enable web search by including "webSearch": {} in the capabilities object of your API request.

A step-by-step reasoning for complex problem solving, mathematical reasoning, and logical analysis. Available for select models (support varies by model and compatibility version).

Important: Some models have reasoning always enabled and cannot disable it. For these models:

• The reasoning capability is always active during inference.
• You cannot turn reasoning off via the API.

To determine if a model has always-on reasoning, check the API response when listing models. Models with mandatory reasoning will indicate this in their capability structure.

Extract and process content from web pages for analysis and summarization.

Files are limited to 7MB each (PDF, DOCX, PNG, JPEG, Markdown, HTML, Plain text). You can upload up to 100 files per conversation with a total size limit of 30MB. PDF files are limited to 100 pages total across all PDFs in a conversation.

All models include moderation for inappropriate content, harmful instructions, personal information, copyrighted material, misinformation, sensitive topics, and security threats.

Model descriptions returned by the API are provided in English and may be updated over time to reflect model improvements or capability changes.

If your application requires translated model descriptions, maintain a translation map in your code keyed by model.id, with fallback to the English description from the API for unknown models. This allows new models to work immediately while you add translations at your own pace.

Models scheduled for removal will include a removal field with an ISO 8601 date (e.g., "removal": "2025-11-17T00:00:00.000Z"). When a model is removed, API requests will fail with error code MODEL_NOT_FOUND and the models endpoint will stop returning that particular model.

POST /v1/conversations/my-conversation-123/messages
Content-Type: application/json
Authorization: Bearer <your-token>

{
  "prompt": "Analyze this document and provide insights",
  "model": "agent-1",
  "content": [
    {
      "type": "document",
      "id": "doc-1234567890123"
    }
  ]
}

POST /v1/conversations/my-conversation-123/messages
Content-Type: application/json
Authorization: Bearer <your-token>

{
  "prompt": "Research the latest developments in AI",
  "model": "gpt-4o",
  "capabilities": {
    "webSearch": {},
    "reasoning": {}
  }
}

Get all available models for compatibility version 1:

GET /v1/models/1
Authorization: Bearer <your-token>

Response:

{
  "items": [
    {
      "id": "agent-1",
      "name": "Agent",
      "provider": "Agent",
      "description": "Automatically selects the best model for speed, quality, and cost",
      "allowed": true,
      "capabilities": {
        "webSearch": {
          "enabled": true,
          "allowed": true
        },
        "reasoning": {
          "enabled": true,
          "allowed": true
        }
      },
      "limits": {
        "maxPromptLength": 30000,
        "maxConversationLength": 256000,
        "maxFiles": 100,
        "maxFileSize": 7000000,
        "maxTotalFileSize": 30000000,
        "maxTotalPdfFilePages": 100
      }
    }
  ]
}

For complete documentation on model endpoints, compatibility versions, and capability schemas, see:

• Models API Reference – Full documentation for model listing and configuration.
• Complete API Documentation – Interactive API reference with all CKEditor AI endpoints.

• Conversations – Use models in interactive AI discussions.
• Reviews – Apply models to content analysis and improvement.
• Actions – Use models for content transformation tasks.