Thinking Machines Releases Inkling Open-Weight Model
Simon Willison · rss · 2026-07-16
Thinking Machines Lab has released its first open-weight model, Inkling.
- It is a Mixture-of-Experts Transformer with 975B total parameters / 41B active parameters.
- Licensed under Apache-2.0, it is positioned as a multimodal foundation model ready for fine-tuning, rather than a frontier-leading model.
- Training data covers 45 trillion tokens, including text, images, audio, and video.
- Inkling-Small (276B total / 12B active parameters) was also teased, though weights will be released after testing.
The author believes this version is a competitive new option in the US open-weight ecosystem, comparable to Chinese open-source models. The post includes examples of the model generating SVGs and describing images, though it misidentified a bird in an image as a "stork or seagull".
More from Models
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11
- 105 hidden bugs, 2 repos: DeepSeek V4.1 Flash fixes 24 at $1.80 vs Opus 5's 27 at $51.33 — ChartsJournalX · 2026-09-11
- awesome-llm-leaderboards: an open-source directory of LLM leaderboards, pricing tables, comparison tools — Last_Establishment_1 · 2026-09-11
- Anthropic claims it works to keep eval environments unidentifiable to models — MaxKannen · 2026-09-11
- Nex N2.5 Pro released on Hugging Face with 407GB of weights — jinnyjuice · 2026-09-11
- RoMa v2 image matching model unveiled in the usual black poster — ducha_aiki · 2026-09-11