{"slug": "mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real", "title": "Mistral's open-source Leanstral 1.5 aces formal math benchmarks and catches real bugs in code", "summary": "Mistral AI released Leanstral 1.5, an open-source model for formal verification in Lean 4, achieving 100% on the miniF2F math benchmark and solving 587 of 672 Putnam problems. The model also caught five real bugs in open-source code, including an overflow bug in the Rust library varinteger.", "body_md": "# Mistral's open-source Leanstral 1.5 aces formal math benchmarks and catches real bugs in code\n\n**Mistral AI released Leanstral 1.5, a free open-source model (Apache 2.0 license) built for formal verification in the Lean 4 programming language.** Lean 4 is designed to formally verify mathematical proofs and software correctness.\n\nMistral says the model hits 100 percent on miniF2F, a formal math benchmark covering problems from high school level up to math olympiad difficulty. On PutnamBench, which includes 672 problems from the Putnam math competition, it solves 587. On the algebra benchmarks FATE-H and FATE-X, which test master's and doctoral-level tasks in areas like group theory and ring theory, it scores top results of 87 and 34 percent.\n\nThe model was trained mainly for math, but Mistral says it also performs well at code verification. In a hands-on test, it scanned 57 open-source repositories and caught five previously unknown bugs, including an overflow bug in the Rust library varinteger. The model is available through [Hugging Face](https://huggingface.co/mistralai/Leanstral-1.5-119B-A6B) and a [free API](https://docs.mistral.ai/models/model-cards/leanstral-1-5). Training involved mid-training, supervised fine-tuning, and reinforcement learning.\n\n```\nAI News Without the Hype – Curated by Humans\n\n\t\t\t\t\tSubscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive \"AI Radar\" frontier report six times a year, full archive access, and access to our comment section.\t\t\t\t\n\n\t\t\t\t\tSubscribe now\n```\n\n[Mistral](https://mistral.ai/news/leanstral-1-5/)", "url": "https://wpnews.pro/news/mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real", "canonical_source": "https://the-decoder.com/mistrals-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real-bugs-in-code/", "published_at": "2026-07-04 07:12:47+00:00", "updated_at": "2026-07-04 07:26:50.716308+00:00", "lang": "en", "topics": ["large-language-models", "ai-research", "ai-products", "ai-tools"], "entities": ["Mistral AI", "Leanstral 1.5", "Lean 4", "miniF2F", "PutnamBench", "FATE-H", "FATE-X", "varinteger"], "alternates": {"html": "https://wpnews.pro/news/mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real", "markdown": "https://wpnews.pro/news/mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real.md", "text": "https://wpnews.pro/news/mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real.txt", "jsonld": "https://wpnews.pro/news/mistral-s-open-source-leanstral-1-5-aces-formal-math-benchmarks-and-catches-real.jsonld"}}