Deep Search API
Build Deeper AI Context.
Multimodal Retrieval API to Enrich Your AI's Context with premium trusted content – RAG, Agents, LLMs.
Proprietary Content
Search paywalled content, academic research, reports across knowledge work domains.
Web
Full-text search across the web beyond simple links. Get back structured content.
Multi-modal
Search across text, images, tables, code, and other data modalities.
Pay-per-use
Only pay for retrievals and usage, with cost effective pricing.
Proprietary Content
Search paywalled content, academic research, reports across knowledge work domains.
🛠️ Built for Engineers, By Engineers
Simple API, Powerful Results
Connect DeepSearch to your LLM apps & agents without rebuilding your stack.
from valyu import Valyu
valyu = Valyu(api_key="<api_key>")
response = valyu.context(
query="Tell me about ancient civilizations",
search_type="proprietary",
max_num_results=3,
max_price=10
)
{
"query": "Tell me about ancient civilizations",
"results": [
{
"content": "Ancient civilizations refer to complex societies that existed before the modern era...",
"relevance_score": 0.95,
"source": "Encyclopedia of World History",
"price": 0.005
},
{
"content": "Some of the most well-known ancient civilizations include Mesopotamia, Ancient Egypt...",
"relevance_score": 0.92,
"source": "Ancient Civilizations Quarterly",
"price": 0.005
},
{
"content": "These early societies developed sophisticated systems of government, religion, and art...",
"relevance_score": 0.88,
"source": "Journal of Archaeological Studies",
"price": 0.005
}
],
"total_price": 0.015
}
Used by engineers, researchers and founders across academia, enterprise and startups
Built For Knowledge Work
Deep research today is only as deep as the open web. DeepSearch API changes that. Power your agents and applications with full-document search, grounded retrieval, and direct answer generation. Built for high-fidelity knowledge tasks across research, analysis, and decision-making.
Grounding
Mitigate against hallucinations by injecting trusted external knowledge into your context windows to anchor outputs to real facts.
Search
Retrieve full text documents, tables, links, and images — structured, controllable, and low-latency.
Reasoning
Improve test-time accuracy and efficiency with context-relevant data retrieval.
Get Answers
Get an LLM answer to a question informed by Valyu's Deep Search API.
Grounding
Mitigate against hallucinations by injecting trusted external knowledge into your context windows to anchor outputs to real facts.
Search
Retrieve full text documents, tables, links, and images — structured, controllable, and low-latency.
Reasoning
Improve test-time accuracy and efficiency with context-relevant data retrieval.
Get Answers
Get an LLM answer to a question informed by Valyu's Deep Search API.
300ms
Latency
5 QPS
Per API Key
99.7%
Up time
Works Where You Work
Connect DeepSearch to your LLM apps & agents without rebuilding your stack.
Frequently Asked Questions
Build Faster with Deeper Context.
DeepSearch gives you clean, relevant results from trusted sources. One API. No scraping. No noise.