From post-training to production: How TinyFish runs its web-native model on Databricks AI Runtime

TinyFish, which builds infrastructure for enterprises that need agents to operate across the live web, is now running its web-native model on Databricks AI Runtime, taking the model from post-training into production.
The company's product spans Search, Fetch and Browser capabilities, aimed at agents that interact with live web content rather than static datasets.
Why it matters: enterprises building web-facing agents get another reference point for pairing a domain-specific model with a managed serving stack, and Databricks positions its AI Runtime as the place where that post-training-to-production transition happens.
Based on reporting from the original publisher. Visit the source for full context and later updates.
Publisher excerpt
TinyFish builds infrastructure for enterprises that need agents to work across the live web. Its Search, Fetch, Browser...