Inkling
#3 in Open-Source LLMsThinking Machines Lab · 21× · last seen Jul 17, 2026
Inkling is the first in-house developed language model by Thinking Machines Lab (founded by former OpenAI CTO Mira Murati), released on July 15, 2026. It is a multimodal Mixture-of-Experts model with 975 billion total parameters, of which only 41 billion are active per Token, and a context window of up to 1 million Token. The model natively processes text, image, and audio, but generates only text outputs, and is available under Apache-2.0 license with open weights via Hugging Face as well as through the Tinker platform and third-party APIs. Thinking Machines explicitly positions Inkling not as the most powerful model, but as an adaptable, cost-efficient foundation for Fine-Tuning via its proprietary Tinker platform.