lean-dojo/LeanCopilot
LLMs as Copilots for Theorem Proving in Lean
Integrates natively into Lean 4 via FFI with CTranslate2 for local inference, while supporting remote model execution through configurable APIs. Provides three core tactics—`suggest_tactics` for LLM-generated tactic proposals, `search_proof` for multi-tactic proof search combined with aesop, and `select_premises` for retrieval-based premise selection—all backed by pre-trained models from LeanDojo or custom user-defined models.
1,244 stars.
Stars
1,244
Forks
122
Language
C++
License
MIT
Category
Last pushed
Feb 17, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/lean-dojo/LeanCopilot"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
ai4co/reevo
[NeurIPS 2024] ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution
SALT-NLP/collaborative-gym
Framework and toolkits for building and evaluating collaborative agents that can work together...
Gen-Verse/LatentMAS
Latent Collaboration in Multi-Agent Systems
sethkarten/LLM-Economist
Official repository of the 2025 paper, LLM Economist: Large Population Models and Mechanism...
WooooDyy/AgentGym-RL
Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon...