Grass has launched a Contents API that converts public web pages into HTML, Markdown, or plain text for AI agents and other model-driven applications.
The official 8 October announcement says the service uses the Grass Network’s bandwidth infrastructure, with fallback paths ranging from lightweight retrieval to full browser rendering. Access is reviewed rather than automatically open.
Grass says an internal 100-site benchmark retrieved 97% of tested URLs, compared with 84% for Firecrawl, 76% for Parallel, and 66% for Exa. On the 50 URLs all four providers retrieved, Grass reported median client latency of 2.08 seconds. The test was designed and published by Grass, not an independent laboratory.
The product creates a direct use case for the project’s DePIN network beyond selling historical training data. AI agents require current information from changing websites, and the API attempts to turn distributed residential bandwidth into a commercial retrieval layer.
TOKEN RECON ASSESSMENT
This clears the AI Crypto threshold because a functioning network has shipped an agent-facing product tied to its decentralized infrastructure. The benchmark is useful but self-reported. The stronger proof will be independent reproduction, paying customers, request volume, contributor economics, privacy controls, and evidence that the system respects site policies while retrieving public data.
