Add model card

This PR adds a model card for the OctoThinker model, linking it to the paper [OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling](https://huggingface.co/papers/2506.20512).

README.md

@@ -0,0 +1,11 @@
+---
+license: cc-by-nc-4.0
+library_name: transformers
+pipeline_tag: text-generation
+---
+
+This repository contains the OctoThinker model from the paper [OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling](https://huggingface.co/papers/2506.20512).
+
+Code: https://github.com/AI4Math/OctoThinker
+
+Project Page: https://huggingface.co/OctoThinker