on hugging science: kimina-prover ✍️
rl-trained model that writes formal lean 4 proofs for olympiad-level maths. machine-checkable, not just "showing working". 7b distilled preview, open on the hub.
huggingface.co/AI-MO/Kimina-P…
rl-trained model that writes formal lean 4 proofs for olympiad-level maths. machine-checkable, not just "showing working". 7b distilled preview, open on the hub.
huggingface.co/AI-MO/Kimina-P…