--- format: "aidr-story-markdown/v1" id: "4cb93994a57ddb2a6fb469642c76479fdbb2f181d1e3e47b24090f7ab29be90d" canonical_url: "https://aidr.today/4cb93994?lang=en" title: "Nvidia's Medical Fine-Tuning Cuts Nemotron 3.5 Transcription Error Rate by 45.8%" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-30T00:23:01.000Z" category: "Models" topics: ["nvidia","llm"] source_urls: ["https://marketbrief.now/ai/nvidias-medical-fine-tuning-cuts-nemotron-35-transcription-error-rate-by-805511be","https://huggingnews.com/ai/nvidias-medical-fine-tuning-cuts-nemotron-35-transcription-error-rate-by-805511be"] summary: "Nvidia's medical fine-tuning of its Nemotron 3.5 Transcribe model reduced the word error rate on medical recordings from 22.44% to 12.17%, a 45.8% improvement achieved using about 14 GPU hours on a single GB300 processor. The company is publishing the fine-tuning recipe and model weights to help developers adapt the tool to specialized vocabularies. Nemotron 3.5 Transcribe is a 30B parameter Mixture-of-Experts model supporting 40 languages and speaker identification, available for commercial use under the OpenMDW-1.1 license." --- # Nvidia's Medical Fine\-Tuning Cuts Nemotron 3\.5 Transcription Error Rate by 45\.8% > [Open the canonical story]() **Published:** 2026-10-30T00:23:01.000Z **Category:** Models **Topics:** nvidia, llm ## Summary Nvidia's medical fine\-tuning of its Nemotron 3\.5 Transcribe model reduced the word error rate on medical recordings from 22\.44% to 12\.17%, a 45\.8% improvement achieved using about 14 GPU hours on a single GB300 processor\. The company is publishing the fine\-tuning recipe and model weights to help developers adapt the tool to specialized vocabularies\. Nemotron 3\.5 Transcribe is a 30B parameter Mixture\-of\-Experts model supporting 40 languages and speaker identification, available for commercial use under the OpenMDW\-1\.1 license\. ## Sources - [Story source]() - [Story source]()