---
title: "Search-o1: Agentic Search-Enhanced Large Reasoning Models"
license: CC-BY-4.0
license_url: https://creativecommons.org/licenses/by/4.0/
license_note: 'paper is arXiv non-exclusive distribution license (verified on arXiv abs page 2026-11-12); this pointer note is our own words and is CC BY 4.0; we link to the paper rather than reproduce it'
source: https://iinventedlight.com/library/notes/2501-05366-search-o1.md
source_license: "arXiv non-exclusive distribution license"
attribution: 'Pointer by iinventedlight (AI-assisted). Source: Xiaoxi Li, Guanting Dong, Jiajie Jin, and 5 coauthors, Search-o1: Agentic Search-Enhanced Large Reasoning Models (arXiv non-exclusive distribution license). This note links to the paper rather than reproducing it.'
author: 'iinventedlight (AI-assisted)'
date: 2026-11-12
status: published
approved_by: "Cole Withington, 2026-11-12 14:44 PT, Librarian chat widget t51s316: \"Yes, publish all 100 second-list arXiv notes\""
checked_on: 2026-11-12
as_of: 2026-11-12
confidence: medium
untested: 'Pointer note only. Paper not reproduced. Abs page and license checked 2026-11-12.'
category: arxiv
topic_area: 'ai-research: agents'
tags: arxiv, agents, rag, reasoning, search-o1
arxiv_id: 2501.05366
alias: 2501-05366-search-o1
type: pointer
format: pointer
sources:
  - id: S1
    title: "Search-o1: Agentic Search-Enhanced Large Reasoning Models"
    publisher: arXiv
    authors: "Xiaoxi Li and 7 coauthors (listed on the arXiv abs page)"
    url: https://arxiv.org/abs/2501.05366
    version_read: "v1 (9 Jan 2025)"
    accessed: 2026-11-12
    license: "arXiv non-exclusive distribution license"
---
# Search-o1: Agentic Search-Enhanced Large Reasoning Models

## Summary
[Search-o1: Agentic Search-Enhanced Large Reasoning Models](https://arxiv.org/abs/2501.05366) (2025) adds agentic search to large reasoning models so they can look things up during an extended thinking trace, instead of relying only on what is already in the weights.

What it introduced: a workflow that triggers retrieval when the reasoner hits an uncertain knowledge point, plus a Reason-in-Documents module that analyzes retrieved text before it is injected back into the chain, to cut noise and keep the trace coherent. The authors evaluate on science, math, and coding reasoning, and on six open-domain question-answering benchmarks.

What it led to: a clear recipe for combining long-chain reasoners with retrieval mid-thought, rather than only retrieving once at the start. Live notes: [2005-11401-retrieval-augmented-generation](https://iinventedlight.com/library/notes/2005-11401-retrieval-augmented-generation/), [2501-12948-deepseek-r1](https://iinventedlight.com/library/notes/2501-12948-deepseek-r1/), and [2310-11511-self-rag](https://iinventedlight.com/library/notes/2310-11511-self-rag/). FLARE, an earlier active-retrieval method, is also a pointer note in this batch.

Where to read it: https://arxiv.org/abs/2501.05366 (v1, 9 Jan 2025). Code: https://github.com/sunnynexus/Search-o1.

The paper is posted on arXiv under the arXiv non-exclusive distribution license. This note links to it rather than reproducing it. The note itself is CC BY 4.0.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Pointer by iinventedlight (AI-assisted). Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it.

---

Agents: [/agents.md](/agents.md) | [/.well-known/agent-library.json](/.well-known/agent-library.json).
