https://inkling2small.com/news/ Announcements about Inkling 2 Small and changes detected in its specifications, pricing and release status. Unofficial. en Fri, 06 Nov 2026 11:44:11 GMT inkling2small.com static build https://inkling2small.com/news/inkling-2-small/ https://inkling2small.com/news/inkling-2-small/ Thu, 05 Nov 2026 00:00:00 GMT Try on Tinker Model card Today we are introducing our second model generation with Inkling 2 Small. It is a Mixture-of-Experts model with 594B total parameters, 17B active, and supports image inputs and variable effort. Inkling 2 Small far surpasses our previous models’ scores on the Artificial Analysis Intelligence Index with a score of 31, reflecting gains in performance on agentic coding, visio https://thinkingmachines.ai/news/one-year-of-tinker/ https://thinkingmachines.ai/news/one-year-of-tinker/ Thu, 15 Oct 2026 00:00:00 GMT One Year of Tinker: Turning Your Expertise into Custom Models https://thinkingmachines.ai/news/putting-task-expertise-into-rl/ https://thinkingmachines.ai/news/putting-task-expertise-into-rl/ Thu, 27 Aug 2026 00:00:00 GMT Putting Task Expertise into RL Achieves State-of-the-Art Performance on Text-to-SQL https://thinkingmachines.ai/news/safety-research-grants/ https://thinkingmachines.ai/news/safety-research-grants/ Mon, 24 Aug 2026 00:00:00 GMT Announcing Safety Research Grants https://inkling2small.com/news/blog-a-safe-path-to-open-weights/ https://inkling2small.com/news/blog-a-safe-path-to-open-weights/ Fri, 31 Jul 2026 00:00:00 GMT Abstract: Safe open-weight models are public goods, as they put AI development and safety work in many hands and make training choices inspectable. Open models also carry real misuse risks, and release is irreversible. To release safely, we must consider both the model and the ecosystem it enters. F https://inkling2small.com/news/inkling-small/ https://inkling2small.com/news/inkling-small/ Thu, 30 Jul 2026 00:00:00 GMT Try on Tinker Model card Hugging Face Today, we are releasing Inkling-Small, an efficient open-weights model that achieves comparable performance to Inkling at a quarter of its size. Inkling-Small is a Mixture-of-Experts transformer with 276B total parameters, 12B active, trained on NVIDIA GB300 NVL72 systems. Like Inkling, it features native reasoning over audio and images, variable thinking effo https://inkling2small.com/news/introducing-inkling/ https://inkling2small.com/news/introducing-inkling/ Wed, 15 Jul 2026 00:00:00 GMT Hugging Face Our mission is to build AI that extends human will and judgment. We have developed a platform that lets anyone customize models, previewed an AI system built for interactive collaboration, and published novel research . Today we are advancing our mission by releasing a model we trained from scratch with the full weights available, so that people can make it their own. Our model, calle https://inkling2small.com/news/blog-the-future-worth-building-is-human/ https://inkling2small.com/news/blog-the-future-worth-building-is-human/ Fri, 10 Jul 2026 00:00:00 GMT The mission of Thinking Machines is to build AI that extends human will and judgment. Artificial intelligence can do more every day, but deciding what it should do is up to us: individuals, organizations, humanity as a whole. These decisions require knowledge and judgment that people acquire through https://thinkingmachines.ai/news/learning-to-replicate-expert-judgment-in-financial-tasks/ https://thinkingmachines.ai/news/learning-to-replicate-expert-judgment-in-financial-tasks/ Tue, 30 Jun 2026 00:00:00 GMT Learning to Replicate Expert Judgment in Financial Tasks https://thinkingmachines.ai/news/interactivity-research-grants/ https://thinkingmachines.ai/news/interactivity-research-grants/ Tue, 19 May 2026 00:00:00 GMT Announcing Interactivity Research Grants https://inkling2small.com/news/blog-interaction-models/ https://inkling2small.com/news/blog-interaction-models/ Mon, 11 May 2026 00:00:00 GMT Today, we’re announcing a research preview of interaction models: models that handle interaction natively rather than through external scaffolding. We think interactivity should scale alongside intelligence; the way we work with AI should not be treated as an afterthought. Interaction models let peo https://thinkingmachines.ai/news/training-llms-to-predict-world-events/ https://thinkingmachines.ai/news/training-llms-to-predict-world-events/ Thu, 19 Mar 2026 00:00:00 GMT Training LLMs to Predict World Events (Guest Post with Mantic) https://thinkingmachines.ai/news/nvidia-partnership/ https://thinkingmachines.ai/news/nvidia-partnership/ Tue, 10 Mar 2026 00:00:00 GMT Thinking Machines Lab and NVIDIA Announce Long-Term Gigawatt-Scale Strategic Partnership https://thinkingmachines.ai/news/tinker-general-availability/ https://thinkingmachines.ai/news/tinker-general-availability/ Fri, 12 Dec 2025 00:00:00 GMT Tinker: General Availability and Vision Input https://thinkingmachines.ai/news/call-for-community-projects/ https://thinkingmachines.ai/news/call-for-community-projects/ Fri, 07 Nov 2025 00:00:00 GMT Tinker: Call for Community Projects https://thinkingmachines.ai/news/tinker-research-and-teaching-grants/ https://thinkingmachines.ai/news/tinker-research-and-teaching-grants/ Wed, 29 Oct 2025 00:00:00 GMT Tinker: Announcing Research and Teaching Grants https://inkling2small.com/news/blog-on-policy-distillation/ https://inkling2small.com/news/blog-on-policy-distillation/ Mon, 27 Oct 2025 00:00:00 GMT LLMs are capable of expert performance in focused domains, a result of several capabilities stacked together: perception of input, knowledge retrieval, plan selection, and reliable execution. This requires a stack of training approaches, which we can divide into three broad stages: Pre-training teac https://thinkingmachines.ai/news/announcing-tinker/ https://thinkingmachines.ai/news/announcing-tinker/ Wed, 01 Oct 2025 00:00:00 GMT Announcing Tinker https://inkling2small.com/news/blog-lora/ https://inkling2small.com/news/blog-lora/ Mon, 29 Sep 2025 00:00:00 GMT Today’s leading language models contain upwards of a trillion parameters, pretrained on tens of trillions of tokens. Base model performance keeps improving with scale, as these trillions are necessary for learning and representing all the patterns in written-down human knowledge. In contrast, post-t https://inkling2small.com/news/blog-modular-manifolds/ https://inkling2small.com/news/blog-modular-manifolds/ Fri, 26 Sep 2025 00:00:00 GMT When we train large neural networks, we need to keep them healthy. We do not want the tensors in the network—either the weights, activations, or gradients—to grow too large or too small. Very small and very large tensors cause a variety of problems not just limited to numerical underflow and overflo https://inkling2small.com/news/blog-defeating-nondeterminism-in-llm-inference/ https://inkling2small.com/news/blog-defeating-nondeterminism-in-llm-inference/ Wed, 10 Sep 2025 07:01:00 GMT Reproducibility is a bedrock of scientific progress. However, it’s remarkably difficult to get reproducible results out of large language models. For example, you might observe that asking ChatGPT the same question multiple times provides different results. This by itself is not surprising, since ge