Thinking Machines Lab Releases Inkling: A 975B-Parameter Open-Weights Multimodal MoE With 41B Active Parameters And Controllable Thinking Effort

Thinking Machines Lab released Inkling, an open-weights multimodal Mixture-of-Experts model with 975B total and 41B active parameters, trained on 45 trillion tokens and supporting up to 1M-token context. The model offers controllable reasoning effort, competitive performance on benchmarks, and requires substantial VRAM for deployment, with fine-tuning available via Tinker. Use cases include voice-and-vision agents, cost-tiered pipelines, and domain-specific fine-tuning.