--- id: "2026-07-15-thinking-machines-inkling" url: "https://postcutoff.com/e/2026-07-15-thinking-machines-inkling/" as_of: "2026-11-19T22:11:00+01:00" date: "2026-07-15" date_precision: day category: open-source importance: 4 confidence: high status: [Confirmed] sources: 5 editor: human editor human_review: null version: "2026-10-04" --- As of: 2026-11-19 22:11 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/e/2026-07-15-thinking-machines-inkling/ # Thinking Machines Lab releases Inkling, its first open-weights model Full title: Thinking Machines Lab releases Inkling, its first open-weights model (975B MoE) Mira Murati's Thinking Machines Lab released Inkling on 2026-07-15: a 975B-parameter (41B active) natively multimodal MoE trained on 45T tokens, with 1M context, under Apache 2.0, plus a preview Inkling-Small (276B / 12B active), positioned for customization via its Tinker fine-tuning platform. ## Key facts - 975B total / 41B active parameters; 45T training tokens across text, images, audio, video; 1M context - Benchmarks (effort=0.99): HLE with tools 46.0%, AIME 2026 97.1%, SWE-bench Verified 77.6%, GPQA Diamond 87.2% - Safety: 78.0% FORTRESS, 98.6% StrongREJECT - Inkling-Small preview: 276B total / 12B active - License Apache 2.0; available on Hugging Face, Tinker, Together, Fireworks, Modal, Databricks, Baseten - ARC Prize: Inkling 36.5% on ARC-AGI-2; Inkling Small 40.1% ## What happened Thinking Machines Lab, founded by former OpenAI CTO Mira Murati, shipped its first broadly usable model as fully open weights under Apache 2.0. Inkling is a sparse MoE with native text/image/audio reasoning and controllable thinking effort, distributed through major inference providers and the company's own Tinker fine-tuning service. ## Why it matters It is the most capable permissively licensed (Apache 2.0) US-origin open model at release, giving Western developers a counterweight to Chinese open-weights leaders, and it underpins Thinking Machines' bet that customers want to own and fine-tune their models. ## Your AI and this story - GPT-6.2 Astra (training cutoff May 2026): 45 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 15 days after its cutoff - Gemini 4 Hydrogen (training cutoff March 2026): 106 days after its cutoff - Grok 4.7 (training cutoff May 2026): 45 days after its cutoff ## Sources 1. [Thinking Machines: Inkling, our open-weights model](https://thinkingmachines.ai/news/introducing-inkling/) (thinkingmachines.ai, official) 2. [Inkling model card](https://thinkingmachines.ai/model-card/inkling/) (thinkingmachines.ai, docs) 3. [Hugging Face blog: Welcome Inkling](https://huggingface.co/blog/thinkingmachines-inkling) (huggingface.co, official) 4. [TechCrunch: Thinking Machines' first open model, Inkling](https://techcrunch.com/2026/07/15/thinking-machines-amps-up-its-bet-against-one-size-fits-all-ai-with-its-first-open-model-inkling/) (techcrunch.com, press) 5. [Simon Willison on Inkling](https://simonwillison.net/2026/Jul/16/inkling/) (simonwillison.net, discussion) ## Changes - 2026-09-29 (filed): Created ## Related - 2026-07-16: [Moonshot AI releases Kimi K3, a 2.8T-parameter open-weights multimodal model](https://postcutoff.com/e/2026-07-16-moonshot-kimi-k3/index.md) - People: [Mira Murati](https://postcutoff.com/person/mira-murati/), [Simon Willison](https://postcutoff.com/person/simon-willison/)