{"version":1,"type":"story","url":"https://digestai.news/story/laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin","json":"https://digestai.news/story/laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin.json","markdown":"https://digestai.news/story/laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin.md","slug":"laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin","headline":"Laya, a 322M-parameter decision engine, gains 26,639 GitHub stars in nine days","summary":"Laya, an open-source decision engine released on September 18, 2026, has rapidly gained traction with 26,639 GitHub stars and 2,326 forks within nine days. Developed by NandhaKishorM and licensed under Apache-2.0, the tool offers a non-autoregressive alternative to using large language models for simple classification tasks. Instead of generating text tokens, Laya performs a single forward pass to answer typed questions, such as multiple-choice labels, intensity scores, or yes/no probabilities, reporting zero output tokens.\n\nThe project ships three checkpoints on Hugging Face, including a 421M-parameter English model and a 322M-parameter multilingual model supporting over 100 languages. The repository claims a latency of 33ms per question on a T4 GPU, dropping to 7.2ms when batched. A hands-on test on a CPU-only VM confirmed that the system runs locally without API keys or GPU requirements, though CPU inference is slower at roughly 0.7 seconds per question. The tool includes a built-in abstention mechanism that flags low-confidence answers, allowing developers to route uncertain cases to human review or larger models.\n\nWhile the speed and cost efficiency are significant, the documentation warns that the shipped checkpoints are over-confident and require users to fit temperature parameters on their own data before relying on the probability scores. Laya is positioned for high-volume, bounded decisions like ticket triage and moderation, where deterministic, fast responses are preferred over the nuanced reasoning of frontier models.","keyPoints":["Laya gained 26,639 GitHub stars and 2,326 forks in nine days since its September 18, 2026 release.","The 322M-parameter multilingual model answers typed questions in one forward pass with zero output tokens.","Docs warn checkpoints are over-confident; users must fit temperatures on held-out data to trust probabilities."],"whyItMatters":"Laya offers a cost-effective, low-latency alternative to LLMs for high-volume classification tasks. By eliminating token generation and API costs, it enables scalable, offline decision-making for support triage and moderation, reducing reliance on expensive frontier models for simple judgments.","category":{"slug":"research","name":"Research","url":"https://digestai.news/category/research"},"entities":{"companies":["Hugging Face"],"models":["Laya","ModernBERT-large"],"people":["NandhaKishorM"]},"firstPublishedAt":"2026-09-27T23:55:59Z","updatedAt":"2026-09-27T23:55:59Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"aifrontierpost.com","title":"Laya: replace LLM-as-a-judge with a 322M-parameter decision engine (26,639 stars in 9 days, hands-on test)","url":"https://aifrontierpost.com/articles/laya-typed-decisions-tutorial","publishedAt":"2026-09-27T23:55:59Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[{"site":"Reddit","url":"https://www.reddit.com/r/LocalLLaMA/comments/1wrzjib/laya_replace_llmasajudge_with_a_322mparameter/","points":null}],"thread":null,"cite":{"text":"Digest AI, \"Laya, a 322M-parameter decision engine, gains 26,639 GitHub stars in nine days\", 27 September 2026, https://digestai.news/story/laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin","publisher":"Digest AI","title":"Laya, a 322M-parameter decision engine, gains 26,639 GitHub stars in nine days","datePublished":"2026-09-27T23:55:59Z","url":"https://digestai.news/story/laya-a-322m-parameter-decision-engine-gains-26-639-github-stars-in-nin"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}