Skip to main content
Date: February 2, 2026 This release includes several improvements to give you more control over content extraction and search filtering.

​
Highlights: maxCharacters Replaces numSentences

The maxCharacters parameter is now the preferred way to control highlight length. The legacy numSentences and highlightsPerUrl parameters are deprecated and will be removed in a future release. During the transition period, numSentences is mapped to a character budget (1 sentence ≈ 1333 characters) and highlightsPerUrl is accepted but ignored. Migrate to maxCharacters for precise control. 
result = exa.search_and_contents(
    "latest AI research",
    highlights={"max_characters": 2000}
)

​
Content Freshness: maxAgeHours

The new maxAgeHours parameter in /contents gives you fine-grained control over content freshness:
  • maxAgeHours: 0 - Always livecrawl for fresh content
  • maxAgeHours: -1 - Cache only, never livecrawl
  • maxAgeHours: 24 - Use cache if content is less than 24 hours old, otherwise livecrawl
This replaces the boolean livecrawl options with more precise age-based control. 
# Get content no older than 24 hours
result = exa.get_contents(
  urls = ["tesla.com"],
  max_age_hours = 24,
  text = True
)

​
Exa MCP Free Tier Limits

Unauthenticated users can now try the Exa MCP for free with the following limits:
  • 3 QPS rate limit
  • 150 free calls per day
To exceed these limits, add your API key to unlock full access.

​
Need Help?

If you have any questions about these updates, please reach out to hello@exa.ai.