Tool University of Rochester
ChemCrow
LLM chemistry package integrating expert tools for reasoning-intensive chemical tasks.[3]
L3Multi-step tool-using agentic workflow (workflow level, not a quality rating) Source: Repository-documented Outcome: Not empirically evaluated here
Research applicability
LLM chemistry package integrating expert tools for reasoning-intensive chemical tasks.[3]
Getting started
Clone the public repository, configure model and tool credentials (including RXN4Chem for reaction tools), and review outputs with a chemist before acting on them.
Maintenance
Retrieved repository showed latest commit dated 2024-12-19; service dependencies may have changed.
License
MIT repository license; re-check dependency and data licences before adoption; no independent license audit performed.
Workflow & human oversight
L3 reflects the documented multi-step tool-using design; humans supervise, safety-check and validate results.
Workflow level is not an evidence score; higher is not better evidence.
Approach
An LLM agent composed with chemistry tools (the default reaction tooling requires RXN4Chem access) to plan and reason over chemical tasks.[3]
Limitations
Repository documentation checked, not experimental runs; reaction suggestions need expert safety and feasibility assessment. No software execution or independent performance validation.
Reproducibility
Stored primary-source extract research-wide/candidate-evidence-007.md, accessed 2026-10-15, read for this review. No software, model or claimed outcome reproduced.
Adoption requirements
Python and model API credentials; default reaction tools require RXN4Chem access.
Comments
No comments yet.
Add a comment
Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.