Yichuan Deng

📢📢📢 Releasing OpenThinker3-1.5B, the top-performing SFT-only model at the 1B scale! 🚀 OpenThinker3-1.5B is a smaller version of our previous 7B model, trained on the same OpenThoughts3-1.2M dataset.

Announcing OpenThinker3-7B, the new SOTA open-data 7B reasoning model: improving over DeepSeek-R1-Distill-Qwen-7B by 33% on average over code, science, and math evals. We also release our dataset, OpenThoughts3-1.2M, which is the best open reasoning dataset across all data scales. Full details are in our ✨new paper✨ - below we share the highlights: BTW, it also works on non-Qwen models😉 (1/N)

Announcing OpenThinker-32B: the best open-data reasoning model distilled from DeepSeek-R1. Our results show that large, carefully curated datasets with verified R1 annotations produce SoTA reasoning models. Our 32B model outperforms all 32B models including DeepSeek-R1-Distill-Qwen-32B (a closed data model) in MATH500 and GPQA Diamond, and shows similar performance in other benchmarks. (1/n)

Want to evaluate your models on reasoning benchmarks? We have integrated many math and coding benchmarks into Evalchemy: AIME24, AMC23, MATH500, LiveCodeBench, GPQA, HumanEvalPlus, MBPPPlus, BigCodeBench, MultiPL-E, and CRUXEval. Further, Evalchemy now supports vLLM and OpenAI, accelerating evals for faster results. We have also added Curator into Evalchemy so you can now evaluate any API based model quickly and reliably. Just add --model curator.

Introducing Evalchemy: your one-step solution for Language Model evaluation!