SID AI · Agentic Search Model

SID-2 and SID-2 Swarm

SID-2 is a small agentic search model that searches a user's data, reads results and returns ranked document IDs, replacing fixed RAG pipelines. SID-2 Swarm runs several agents that explore different parts of the corpus in parallel. On SID's own 506-question benchmark across 19 corpora, SID-2 Medium scores 0.585 NDCG in 3.11 s, and the 12-agent Swarm scores 0.720, with all figures self-reported. It is available through SID's hosted API and Baseten, with Python, TypeScript and Go SDKs, but parameters and weights are undisclosed.

Open the source ↗

Similar Agents & Automation releases

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.