Skip to main content
Our API pricing includes Search, Research Agent, and processing/indexing of user-provided content.
API Service   Description   Price
Search   Fast mode (per query unit)   $0.015
   Smart mode (per query unit)   $0.03
   Batch search (per query unit)   $0.0075
Research Agent   Input Tokens (per million)   $3.00
   Output Tokens (per million)   $9.00
   Web Search (per thousand)   $30.00
Content Processing & Indexing   TXT uploads (per thousand pgs)   $12.00 per year
   PDF uploads (per thousand pgs)   $42.00 per year
🚀 Access to the API requires credentials. - Request API Access
The Research Agent is a model endpoint designed for conversational intelligence. It supports multi-turn dialogue, allowing users to ask follow-up questions and refine their queries. Built on real-time retrieval-augmented generation (RAG), it combines an LLM framework with search capabilities across proprietary, premium, and web content sources.The result is a synthesized, context-aware response that reflects reasoning, summarization, and insight generation, ideal for business/financial analysis and complex decision support.In contrast, the Search Service is a retrieval-only API. It indexes structured and unstructured data to return the most relevant documents or snippets, without interpretation or synthesis. It’s optimized for low latency and high precision, making it ideal for tasks such as grounding LLMs, populating context windows, powering search tools, or feeding downstream agents with raw but relevant content.
  • Research Agent = LLM + RAG → contextual, reasoned answers.
  • Search = fast, relevant retrieval → raw source material.
Fast mode gives you full control over all filters for precise, deterministic, low-latency results. Smart mode handles query understanding for you, automatically inferring intent, applying filters, and running sub-queries for broader coverage. Ideal for natural-language questions without pre-processing. Higher latency than Fast mode.
Request ParameterFast ModeSmart Mode
Text✅✅
Temporal filter✅ Manual or Partially Automated✅ Manual or Fully Automated
Source filter✅✅ Manual or Fully Automated
Reporting date filter✅ Fully Automated
Document type filter✅ Fully Automated
Entity filter✅ Manual or Partially Automated Fully Automated
Reporting entity filter✅ Fully Automated
Keyword filter✅ Fully Automated
Sentiment filter✅ Fully Automated
Ranking parameters✅ Fully Automated

​
Search Service Pricing

This simple, pay-per-query model allows predictable and scalable usage across applications. Fast Mode: $0.015 per query unit
  • Each query unit includes up to 10 search results from the Bigdata Vector Database (Chunks of text)
  • No additional charges for retrieval of Knowledge Graph data
Smart Mode: $0.03 per query unit
  • Analyzes the query text to automatically define search filters and runs multiple sub-queries to ensure better coverage
  • Ideal for sending user questions directly without pre-processing
Batch Search: $0.0075 per query unit
  • Asynchronous queries at a 50% discount over fast mode (Smart mode is not supported)
  • Processing time depends on system load. We aim to process all query batches within 24 hours.
You can select the search mode via the search_mode request parameter in the Search Documents endpoint. The MCP tool bigdata_search currently uses only the fast mode.

​
Research Agent Service Pricing

A research request consumes input, and output tokens on the different research steps. The thinking tokens are included as output tokens.
  1. Input Tokens
    • This covers the cost of processing the user’s prompt, along with any context included in the request.
    • Tokens roughly equate to chunks of text (1,000 tokens ≈ 750 words).
  2. Output Tokens
    • This reflects the cost of generating the response from the model and the thinking tokens to plan the next step.
    • The more detailed the answer, the more output tokens are consumed.
  3. Web Search
    • $30.00 per 1,000 performed public web searches

​
Content Processing & Indexing

This pricing covers the annual cost of uploading, processing, and indexing your internal content, such as research reports, manuals, and files so it becomes searchable and usable ONLY by you on Bigdata and through our Search or Research (Agent) services.
  1. TXT Uploads & Storage
    • $12.00 per 1,000 pages/year
    • One TXT page = 3,000 characters of plain text
    • Suitable for structured or text-only documents
  2. PDF Uploads & Storage
    • $42.00 per 1,000 pages/year
    • One PDF page = one actual page, regardless of content density
    • Higher cost reflects advanced parsing: we extract and interpret images, charts, graphs, and tables, making their insights searchable and indexable
Privacy Note:
  • Your data is never viewed by us or used for model training. All content is securely processed and remains private to your account.