{"slug": "murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open", "title": "Murati's 975B Model Fine-Tuned Itself on Launch Day — and Cracked the Top US Open-Weight Spot", "summary": "Mira Murati's Thinking Machines Lab released Inkling, a 975B-parameter open-weight model, on July 15. Within hours, Artificial Analysis scored Inkling at 41 on its Intelligence Index, making it the top-scoring open-weight model ever released by a US lab, three points above NVIDIA's Nemotron 3 Ultra. The launch demo showed Inkling fine-tuning itself via the Tinker API to become a lipogram model that avoids the letter 'e', and the lab openly stated that Inkling is not the strongest overall model available today.", "body_md": "Member-only story\n\n# Murati's 975B Model Fine-Tuned Itself on Launch Day — and Cracked the Top US Open-Weight Spot\n\nThe launch demo wasn’t a benchmark chart. It was a terminal session in which Inkling — the new 975B-parameter model from Mira Murati’s Thinking Machines Lab — wrote its own fine-tuning job, ran it through the company’s Tinker API, evaluated the result, and loaded its own new weights back into the coding harness. The target behavior: become a lipogram model that never uses the letter “e,” something prompting alone can’t reliably achieve. Seven steps later, the model was answering questions with no e’s — because it had post-trained itself.\n\nWithin hours of the July 15 release, Artificial Analysis scored Inkling at 41 on its Intelligence Index — making it the top-scoring open-weight model ever released by a US lab, three points above NVIDIA’s Nemotron 3 Ultra (38) and far above Gemma 4 31B (29) and gpt-oss-120b (24).\n\nAnd here’s the part almost nobody saw coming: the launch post says, in plain text, “Inkling is not the strongest overall model available today, open or closed.” A frontier lab spent 18 months and part of a $2 billion seed round training a model from scratch, then led with a confession.\n\nI spent the weekend reading the technical report, digging through the Hacker News thread, and running the model through Tinker’s playground and…", "url": "https://wpnews.pro/news/murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open", "canonical_source": "https://pub.towardsai.net/muratis-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open-weight-spot-05dd5fac2fcc?source=rss----98111c9905da---4", "published_at": "2026-07-20 06:00:18+00:00", "updated_at": "2026-07-20 13:21:51.309450+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-products", "ai-research", "ai-startups"], "entities": ["Mira Murati", "Thinking Machines Lab", "Inkling", "Tinker API", "Artificial Analysis", "NVIDIA", "Nemotron 3 Ultra", "Gemma 4"], "alternates": {"html": "https://wpnews.pro/news/murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open", "markdown": "https://wpnews.pro/news/murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open.md", "text": "https://wpnews.pro/news/murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open.txt", "jsonld": "https://wpnews.pro/news/murati-s-975b-model-fine-tuned-itself-on-launch-day-and-cracked-the-top-us-open.jsonld"}}