Agentic Search
Mistral's multi-step retrieval layer that lets models search, open, navigate, read, and grep documents iteratively instead of one-shot RAG.
Overview
Agentic Search, announced August 20, 2026, is a retrieval layer built on top of existing search indexes that gives models five tools -- search, open, navigate, read, and grep -- to iteratively find, inspect, and verify information across long or complex documents rather than relying on single-shot RAG retrieval. Mistral reports it triples accuracy on FinanceBench (26.7% to 86% with Mistral Medium 3.5) while cutting token usage by 23.9% and p90 latency from 255s to 154s, and drives a 45.6-point accuracy gain on the table-heavy OfficeQA Pro benchmark. It is built into Mistral's Search Toolkit, Libraries, Studio, and Vibe, and is deployable in cloud or on-premises environments without model fine-tuning.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- Five-tool retrieval loop: search, open, navigate, read, grep
- Works on top of existing search indexes without fine-tuning
- 3x accuracy on FinanceBench (26.7% to 86%) with reduced token use and latency
- +45.6 point accuracy gain on OfficeQA Pro table-heavy multi-document questions
- Integrated into Mistral Search Toolkit, Libraries, Studio, and Vibe; cloud or on-premises
Use cases
- High-accuracy retrieval over financial filings and regulatory documents
- Multi-document, table-heavy question answering
- Enterprise RAG pipelines needing verifiable, multi-step retrieval
Pricing
Not separately priced; delivered as part of Mistral Search Toolkit, Studio, Libraries, and Vibe.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 20 Aug 2026
- Record updated
- 23 Sept 2026
- Record last checked
- 23 Sept 2026