Kimina-Prover-Distill-0.6B is a theorem proving model developed by Project Numina and the Kimi team, focusing on the ability to solve competition-style problems in Lean 4. It is a distilled version of the Kimina-Prover-72B model, achieving an accuracy of 68.85% on MiniF2F-test.

Natural Language ProcessingTransformersEnglish

A theorem proving model developed by the Project Numina and Kimi teams, focusing on the ability to solve competition-style problems in Lean 4.

Scientific ComputingTransformersEnglish

A theorem proving model developed by the Numina project and the Kimi team, focusing on improving the ability to solve competition-style problems in Lean 4.

Natural Language ProcessingTransformersEnglish

Kimina-Prover-Distill-8B is a theorem proving model developed by the Project Numina and Kimi teams, focusing on the ability to solve competition-style problems in Lean 4.

Natural Language ProcessingTransformersEnglish

An open-source large language model specifically designed for Lean 4 formal theorem proving, achieving exceptional accuracy through recursive theorem proving processes and reinforcement learning training.

Natural Language ProcessingTransformersEnglish

An open-source large language model designed for Lean 4 formal theorem proving, collecting data through recursive theorem proving processes, combining informal and formal mathematical reasoning.

Natural Language ProcessingTransformers

Herald is a Lean 4 dataset annotated with natural language, mainly used for research in the fields of natural language processing and formal verification.

Natural Language ProcessingTransformers

BFS-Prover is the most advanced theorem proving system in Lean4, developed based on the Qwen2.5-Math-7B large language model. It can automatically generate tactics according to the proof state in Lean4 and gradually advance the proof process of mathematical theorems.

Natural Language ProcessingTransformersEnglish

InternLM-Step-Prover is an advanced LEAN4 step prover based on a 7B language model. It is trained on Lean-Github and multiple synthetic datasets and performs excellently in mathematical benchmark tests such as MiniF2F, ProofNet, and Putnam, demonstrating strong formal mathematical proof capabilities.

Natural Language ProcessingTransformersEnglish

A particularly satisfying sketch manga style, leaning towards the semi-realistic domain. The 'daiton' tag can be used but is not mandatory.

Computer VisionDiffusers

LeanDojo is a retrieval-augmented language model-based theorem proving system designed to enhance automated theorem proving by combining language models with retrieval techniques.

Scientific ComputingTransformers