Oossa

SID AI launches SID-2 search model with 0.720 NDCG in swarm mode

SID AI says its new SID-2 model reaches 0.720 NDCG using a 12‑agent swarm, outperforming GPT‑6 Astra on the company's benchmark.

NoteBy Published by Oossa: 1 min read

SID AI released SID-2, an agentic search model trained with reinforcement learning. A single SID-2 agent returns results with a 0.585 NDCG score in about 3.1 seconds. When 12 agents work together as a swarm, the score rises to 0.720 NDCG, which SID AI says is higher than GPT‑6 Astra on its own benchmark.

Why it matters

Higher NDCG scores could give developers more accurate search results when they use SID‑2 in multi‑agent applications.

Was this article useful?
Share

Read next

Oossa · Newsletter

The week in AI, explained

Every Monday: the stories worth knowing, in plain language. Free, no spam.