Add model card

README.md

@@ -0,0 +1,20 @@
+---
+language:
+- en
+license: apache-2.0
+tags:
+- lean4
+- theorem-proving
+- verification
+base_model: Qwen/Qwen3-8B
+---
+
+# Goedel-Verifier-Preview-8B
+
+A Lean 4 code verification model fine-tuned from Qwen3-8B through supervised fine-tuning and reinforcement learning with online proof checking.
+
+## Model Details
+
+- **Base model:** Qwen3-8B
+- **Training:** SFT on decomposition and proof completion trajectories, followed by GRPO with online Lean 4 verification rewards
+- **Use case:** Verifying Lean 4 proof attempts for code correctness specifications