{
  "abstract": "Retrosynthetic planning advances through decisions that reshape the remaining chemical problem: a locally plausible disconnection can leave precursors whose chemoselectivity constraints complicate the rest of the route. Existing planners often channel model proposals through search or template procedures, leaving open whether a general-purpose large language model (LLM) can itself sustain and revise route strategy. We developed Rachel, a stateful environment that executes and checks LLM-directed chemistry but prescribes neither a search policy nor a stopping rule. Without supplied reference routes or route-level solutions, GPT-5.5 achieved strict closure for 111 of 120 PaRoutes120 targets and 24 of 25 targets in the separate RF25 difficult-target cohort. RF25 was drawn largely from studies published after GPT-5.5's reported knowledge cutoff. Closure required complete routes and independent source resolution of every terminal precursor after planning. On a shared PaRoutes subset, forward-model support exceeded that of most comparator methods, and Rachel received the highest mean overall route score from both method-blinded LLM evaluators. Recorded trajectories showed continued model-proposed chemistry, with revised strategies carried into subsequent steps. Replacing LLM route decisions with fixed policies reduced strict closure to 6-15/120 despite continued local chemical execution; restricting planning support also reduced closure in RF25. Within Rachel, a general-purpose LLM coordinated successive chemical choices and revised its strategy as earlier decisions reshaped the remaining problems.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.25118",
  "author_total": 6,
  "authors": [
    {
      "name": "Qisheng Li"
    },
    {
      "name": "Shunchao Jiang"
    },
    {
      "name": "Chen Qi"
    },
    {
      "name": "Xin Su"
    },
    {
      "name": "Da Han"
    },
    {
      "name": "Guangyong Chen"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/bwq63v3ymfbdn.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.25118",
  "fulltext_url": null,
  "id": "bwq63v3ymfbdn",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.25118",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.AI"
    ],
    [
      "arxiv",
      "physics.chem-ph"
    ],
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7214114727",
  "pdf_url": "https://arxiv.org/pdf/2609.25118",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/bwq63v3ymfbdn.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:23:39Z",
      "native_id": "2609.25118",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.25118"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "arxiv:physics.chem-ph",
      "subfield": "physical"
    }
  ],
  "title": "Rachel: A general-purpose language model directs and revises retrosynthetic routes",
  "title_text": "Rachel: A general-purpose language model directs and revises retrosynthetic routes",
  "type": "preprint",
  "updated_at": "2026-10-30T12:23:39Z",
  "url": "https://research.draconitic.com/paper/bwq63v3ymfbdn",
  "venue": "arXiv",
  "versions": []
}