DeepSeek Prover V2 Advances Neural Theorem Proving

DeepSeek AI releases DeepSeek-Prover-V2, an open-source LLM for Lean 4 theorem proving using recursive proof search and reinforcement learning, achieving top results on MiniF2F.

One continuously updated timeline instead of dozens of separate articles. New developments are appended as the story evolves.

1. ReleaseApr 30, 2025, 03:46 PM 78%

   DeepSeek AI unveils DeepSeek-Prover-V2, an open-source LLM for Lean 4 theorem proving with recursive proof search and reinforcement learning, outperforming benchmarks.

   DeepSeek AI releases DeepSeek-Prover-V2, an open-source LLM for Lean 4 theorem proving using recursive proof search and reinforcement learning, achieving top results on MiniF2F.

   Read the full story →