APIS FOR AI AGENTS
Search + content extraction in one platform. Web, research, and financial data and clean page conten, in a single call.
AN AGENT IS ONLY AS GOOD AS ITS CONTEXT
Most “search” tools return web links. Agents need sources and content. Blog posts about SEC filings, not the filings. Articles about clinical trials, not the trial data. Pages full of navigation and pop-ups, not the text your model can use.
Discover howWhat are the 5 year historical stock prices for Nvidia and AMD?
Found 2 results
Price of AMD for 2021-01-28 to 2026-01-28
97%Search API
Content API
Search API
Content API
Search, built for AI agents
The highest accuracy search API. Web, research, and financial data in a single call.
Web + Proprietary
Search the open web alongside licensed datasets: arXiv + PubMed, full-text SEC filings, live market data from 70+ exchanges, clinical trials, patents, and economic indicators. One API. Every source your agent needs.
Natural Language Queries
Agents don’t want to parse ticker symbols or look up NCT IDs. They want to ask for what they need. “Phase 3 melanoma immunotherapy trials” works — no database identifiers required.
AI-Ready Outputs
Results come with content, relevance scores, and source metadata. Feed them directly into your agent’s context window. No extraction. No cleaning. No post-processing.
What are the 5 year historical stock prices for Nvidia and AMD?
Stocks
Crypto
Forex
ETFs
Funds
Commodities
Price of AMD for 2021-01-30 to 2026-01-30
"ticker": "AMD",
"period": "5Y",
"data_points": 60,
"prices": [89.2, 112.4, ...]
}
Click to expand • 60 data points
Extract content from any URL
Turn web pages into clean, structured data — Markdown, JSON, screenshots, or AI summaries.
Clean Markdown
Get page content as structured markdown. Headlines, paragraphs, lists, tables — preserved.
Structured Extraction
Define a JSON schema. Extract specific fields from any page.
AI Summaries
Summarize the page with control over length and focus.
Screenshot capture
Need the visual? Get a screenshot of any URL alongside the extracted content.
Tesla, Inc.
Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934
Key Financials
Revenue
$96.77B
Net Income
$7.93B
Total Assets
$106.62B
Employees
140,473
Contents
Part I — Business
15
Part I — Risk Factors
27
Part II — Financial Statements
39
Part II — MD&A
51
Part III — Directors & Officers
63
Use cases:
How users use Valyu
Research pipelines
Extract full articles from search results for deeper analysis
Data enrichment
Pull structured fields from company pages, docs, product listings
Content monitoring
Track changes over time with consistent extraction
Use cases:
How users use Valyu
Research pipelines
Extract full articles from search results for deeper analysis
Data enrichment
Pull structured fields from company pages, docs, product listings
Content monitoring
Track changes over time with consistent extraction
Three lines to power your agent
No pipelines to wire up. No parsing. No cleanup. Three lines is all it takes to give your agent real context.
from valyu import Valyu
valyu = Valyu(api_key="*********************************")
response = valyu.search(
"CRISPR gene therapy clinical trials Phase 3"
)
# Access results
for result in response.results:
print(f"Title: {result.title}")
print(f"URL: {result.url}")
print(f"Content: {result.content}")Why choose Valyu
Intelligent Routing
Valyu analyzes your query and routes to the right sources automatically. Ask about “Tesla earnings” — it hits financial databases. Ask about “mRNA vaccine mechanisms” — it routes to academic literature. Faster. More relevant. Lower cost.
Control & Guardrails
Restrict results by domains, datasets, or URLs. Filter by date range. Set relevance thresholds. Control spend with max_price.
Predictable, transparent pricing
Pay per retrieval. No subscriptions. No token counting. Search API: $0.003 per result (web) | $0.01+ per result (proprietary) Contents API: $0.001 per successful extraction
SoTA across several benchmarks
FreshQA is updated roughly every week, contains 600 time-sensitive questions that assess how effectively APIs handle recent and evolving information. The benchmark is critical for use cases such as news summarization, event tracking, and trend monitoring. Our evaluation used the Aug 27th test set, the most recent release available at the time of testing.
pip install valyuimport { webSearch } from "@valyu/ai-sdk";
const { text } = await generateText({
model: openai('gpt-5'),
tools: { webSearch: webSearch() },
prompt: 'Research the latest AI chip developments'
});
