{"version":1,"type":"story","url":"https://digestai.news/story/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode","json":"https://digestai.news/story/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode.json","markdown":"https://digestai.news/story/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode.md","slug":"supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode","headline":"Supersonic Labs releases Julia 1, a 144.3M-parameter open decision model","summary":"Supersonic Labs, a Brazilian AI lab, has released Julia 1, a compact decision model with 144.3M parameters. Unlike standard chatbots, Julia 1 does not generate text; instead, it accepts context, a question, and 2 to 20 candidate answers to select one option and return probability scores. The model is built on JHU CLSP’s mmBERT-small encoder and is available on Hugging Face under the Apache 2.0 license. It runs locally on CPUs or BF16-capable GPUs, with an ONNX build supporting browser-based inference via WebGPU.\n\nThe lab reports that Julia 1 outperformed TypeSafe’s Jev reference values in three of four benchmark pilots, including AG News and DAIR Emotion. However, it significantly underperformed on the Banking77 dataset, scoring 64/100 compared to Jev’s 87/100 reference. Total cloud GPU training costs were approximately US$104.08. On-device latency tests showed a median of 33.15 ms per decision on an Apple M4 chip, while an Intel Core i5-1235U took 107.83 ms for simpler tasks.\n\nSupersonic Labs notes that Julia 1 is limited to comparing supplied options and cannot handle multi-step calculations or missing facts. A hosted API is planned at $0.025 per million tokens but is not yet open. Julia 2, featuring a proprietary foundation architecture, is currently in development.","keyPoints":["Julia 1 is a 144.3M-parameter model released under Apache 2.0 license by Supersonic Labs.","The model runs on CPUs with a median latency of 33.15 ms on Apple M4 hardware.","Training cost approximately US$104.08 in cloud GPUs, with a planned API price of $0.025/MTok."],"whyItMatters":"Julia 1 offers a low-cost, CPU-friendly alternative for classification tasks, enabling local deployment without expensive GPU infrastructure. Its open weights and low latency make it viable for edge devices and privacy-sensitive applications.","category":{"slug":"models","name":"Generative AI & Models","url":"https://digestai.news/category/models"},"entities":{"companies":["Supersonic Labs","JHU CLSP","TypeSafe","Apple","Samsung","Intel"],"models":["Julia 1","mmBERT-small","Jev"],"people":[]},"firstPublishedAt":"2026-09-26T19:50:57Z","updatedAt":"2026-09-26T19:50:57Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"MarkTechPost","title":"Supersonic Labs Releases Julia 1: A 144.3M-Parameter Open Decision Model That Runs on a CPU","url":"https://marktechpost.com/2026/09/26/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-model-that-runs-on-a-cpu","publishedAt":"2026-09-26T19:50:57Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"Supersonic Labs releases Julia 1, a 144.3M-parameter open decision model\", 26 September 2026, https://digestai.news/story/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode","publisher":"Digest AI","title":"Supersonic Labs releases Julia 1, a 144.3M-parameter open decision model","datePublished":"2026-09-26T19:50:57Z","url":"https://digestai.news/story/supersonic-labs-releases-julia-1-a-144-3m-parameter-open-decision-mode"},"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"}