Thinking Machines Lab Releases Inkling, a 975 Billion Parameter Open Weights AI Model Under Apache 2.0

Thinking Machines Lab released Inkling on Wednesday, an AI model with 975 billion parameters. The company, founded in early 2025 by former OpenAI CTO Mira Murati, is providing the model under the Apache 2.0 license, which allows developers to freely fine-tune and use it.

Inkling is currently the largest open-weights model from the United States and is positioned as an alternative to Chinese open-weights models such as DeepSeek V4, GLM 5.2, and Kimi K2.6. It is available now on Thinking Machines' Tinker platform and via Hugging Face for direct download.

Model Specs, Hardware Requirements, and Positioning Against Chinese Open Weights

Inkling combines a mixture of expert architecture with the following features:

975 billion total parameters

256 routed experts, and two shared experts

Six experts activate for each token, resulting in around 41 billion parameters during inference. It supports a token context window of one million

The model was trained on 45 trillion tokens encompassing text, images, audio, and video and was developed using Nvidia GB300 NVL72 systems

While inspired by DeepSeek-V3, Inkling was trained from scratch by Thinking Machines, rather than fine-tuned from existing weights.

Running Inkling at native 16-bit precision requires more than two terabytes of GPU memory. Practical hardware setups include about eight Nvidia B300 accelerators or sixteen Nvidia H200 accelerators.

For users with less GPU capacity, Thinking Machines offers an NVFP4 quantized version of the model that requires approximately half the GPUs. This version trades some precision for reduced memory usage while maintaining most of the model's capabilities.

Reasoning Capabilities, Apache 2.0 Licensing, and Developer Access

Like most modern frontier models, Inkling is a reasoning model trained with reinforcement learning to use chain-of-thought thinking before responding.

Thinking Machines claims Inkling matches Nvidia's Nemotron 3 Ultra, which was previously the largest Ameri…