> This page is for version 2025-08-08.
> For other versions, use one of these documentation indexes:
> - 2025-11-27 (default): https://docs.caesar.xyz/2025-11-27/llms.txt
> - 2025-08-08: https://docs.caesar.xyz/2025-08-08/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.caesar.xyz/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.caesar.xyz/_mcp/server.

# Caesar API Documentation

![Alt text](file:assets/header.png "Caesar")

The Caesar API gives developers a simple way to turn queries and documents into fully cited, expert-level answers. Instead of stitching together dozens of APIs or building a complex data pipeline, Caesar handles the heavy lifting: scraping, retrieving, reasoning, and synthesizing responses. Every answer comes with the underlying sources so you can surface real-time knowledge in your application without integrating endless third-party tools.

## Why Caesar?

#### Beyond search

Caesar doesn't just find information. It researches, validates, and synthesizes it into coherent, citation-backed insights.

#### Horizontal intelligence

Works across all domains without specialized configuration. From financial analysis to scientific research, Caesar adapts to your query context.

#### Production ready

Async-first architecture with robust error handling, rate limiting, and monitoring. Build responsive applications without blocking on research tasks.

#### Transparent sourcing

Every claim is backed by citations. Caesar provides source URLs, relevance scores, and content snippets for complete transparency.

## Core capabilities

* **Multi-source research**: Searches across the internet and proprietary databases
* **Intelligent synthesis**: Combines findings into coherent, contextual responses
* **Citation tracking**: Links every insight to its source with relevance scoring
* **Iterative refinement**: Uses compute units to validate and expand findings
* **Format transformation**: Applies system prompts to tailor output structure

## Getting started

Caesar operates on an **asynchronous job model**:

1. **Submit a research query** → Receive a job ID
2. **Poll for status** → Check job progress
3. **Retrieve results** → Get synthesized research with citations

```bash
# Submit research request
curl -X POST https://api.caesar.xyz/research \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "Your research question here"}'

# Check status
curl -X GET https://api.caesar.xyz/research/{job_id} \
  -H "Authorization: Bearer YOUR_API_KEY"
```

## Key concepts

#### Compute Units

Control research depth with compute units (1-10). Each unit represents \~1 minute of processing and enables additional reasoning loops. Default is 1 CU.

[Learn more →](/documentation/get-started/compute-units-cu)

#### System Prompts

Transform output format without affecting research logic. System prompts apply post-processing to synthesized content, enabling custom formatting for different use cases.

[Learn more →](/documentation/get-started/prompt-guidance)

#### Pagination

Navigate large result sets efficiently with page-based pagination. All list endpoints support consistent pagination parameters for predictable data retrieval.

[Learn more →](/documentation/get-started/pagination)

#### Rate Limits

5 concurrent jobs, 200 monthly compute units. Limits apply to job creation only, reading results is unlimited.

[Learn more →](/documentation/get-started/rate-limits)

## Support

* **Documentation**: Explore the sidebar for detailed guides
* **API Playground**: Test endpoints in the [API Reference](/api-reference)
* **Contact**: [support@caesar.xyz](mailto:support@caesar.xyz)

## Learn more

#### [API Reference](/api-reference)

Explore all available endpoints

#### [Compute Units](/documentation/get-started/compute-units-cu)

Control research depth

#### [Rate Limits](/documentation/get-started/rate-limits)

Understand usage limits