Qwen2.5-32B and Seed-OSS-36B models finetuned on a EM dataset. Inferece instructions: https://docs.axolotl.ai/docs/inference