{"slug": "vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo", "title": "VibeThinker: 3B param model that beats Opus 4.5 on reasoning with novel SFT+GRPO", "summary": "Researchers developed VibeThinker-3B, a 3-billion-parameter language model that achieves reasoning performance matching or exceeding models orders of magnitude larger, scoring 94.3 on AIME26 and 80.2 on LiveCodeBench v6. The model uses a novel post-training pipeline combining curriculum-based supervised fine-tuning, multi-domain reinforcement learning, and offline self-distillation, demonstrating that compact models can reach frontier-level reasoning without compromising instruction controllability.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 15 Jun 2026]\n\n# Title:VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models\n\n[View PDF](/pdf/2606.16140)\n\n[HTML (experimental)](https://arxiv.org/html/2606.16140v1)\n\nAbstract:This technical report introduces VibeThinker-3B, a compact dense model with 3B parameters developed to investigate how far verifiable reasoning can be pushed within a strictly small-model regime. Building upon the Spectrum-to-Signal post-training paradigm, we systematically enhance the model through an optimized pipeline that includes curriculum-based supervised fine-tuning, multi-domain reinforcement learning, and offline self-distillation. Experimental evaluations demonstrate that VibeThinker-3B achieves frontier-level performance on highly demanding verifiable tasks. Specifically, it attains a score of 94.3 on AIME26 (improving to 97.1 with claim-level test-time scaling), an 80.2 Pass@1 on LiveCodeBench v6, and exhibits strong out-of-distribution generalization with a 96.1\\% acceptance rate on recent unseen LeetCode contests. This effectively places it in the performance band of first-tier reasoning systems, matching or exceeding flagship models that are orders of magnitude larger, such as DeepSeek V3.2, GLM-5, and Gemini 3 Pro. Furthermore, a score of 93.4 on IFEval confirms that this extreme reasoning enhancement does not compromise strict instruction controllability. Extending our previous 1.5B work, these findings motivate the Parametric Compression-Coverage Hypothesis, which views verifiable reasoning as compressible into compact reasoning cores, while open-domain knowledge and general-purpose competence require broad parameter coverage over facts, concepts, and long-tail scenarios. This perspective suggests that compact models are not merely deployment-efficient substitutes, but a complementary path toward frontier-level performance in parameter-dense capability regimes.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo", "canonical_source": "https://arxiv.org/abs/2606.16140", "published_at": "2026-06-23 02:01:25+00:00", "updated_at": "2026-06-24 00:53:23.575388+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "machine-learning", "ai-research"], "entities": ["VibeThinker-3B", "DeepSeek V3.2", "GLM-5", "Gemini 3 Pro", "AIME26", "LiveCodeBench v6", "IFEval", "LeetCode"], "alternates": {"html": "https://wpnews.pro/news/vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo", "markdown": "https://wpnews.pro/news/vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo.md", "text": "https://wpnews.pro/news/vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo.txt", "jsonld": "https://wpnews.pro/news/vibethinker-3b-param-model-that-beats-opus-4-5-on-reasoning-with-novel-sft-grpo.jsonld"}}