{"selector":"[data-hero-shape]","layers":[{"grid":{"columns":8,"rows":5,"x":50,"y":50,"pitch":100},"nodes":[9,11,18,20,27,29],"edges":[[9,18],[11,18],[11,20],[18,27],[20,27],[20,29]]},{"grid":{"columns":8,"rows":5,"x":50,"y":50,"pitch":100},"nodes":[16,25,34],"edges":[[16,25],[25,34]]},{"grid":{"columns":8,"rows":5,"x":50,"y":50,"pitch":100},"nodes":[5,12,14],"edges":[[5,12],[5,14]]},{"grid":{"columns":8,"rows":5,"x":50,"y":50,"pitch":100},"nodes":[39],"edges":[]},{"grid":{"columns":8,"rows":5,"x":50,"y":50,"pitch":100},"nodes":[26,33,35],"edges":[[26,33],[26,35]]}]}

Inkling

Open weights, ready to tinker

Efficient models for text, images, and audio, available to fine-tune on Tinker.

Try on Tinker

The Inkling family

Inkling 2 Small

NEW

Our most capable model, and the first of our second generation, with a big jump in agentic coding and vision.

594B total 17B active

Learn more

Inkling

Our largest model, with native reasoning over text, images, and audio.

975B total 41B active

Learn more

Inkling-Small

Matches its larger sibling on many benchmarks at a quarter of the size, with lower cost and latency.

276B total 12B active

Learn more

Inkling 2 Small is closed weight at release. We intend to release the full weights in the coming weeks, pending the results of additional safety testing.

Built for real work

  • General intelligence

    Performs well across knowledge, math, and science.

  • Agentic coding & tools

    Writes code and uses tools to get work done.

  • Controllable effort

    Adjust thinking time to balance speed and performance.

  • Forecasting & confidence

    Makes predictions with well-calibrated confidence.

  • Instruction following

    Handles detailed, multi-step instructions reliably.

{"inkling":{"grid":3,"nodes":[1,3,7],"edges":[[1,3]]},"tinker":{"bays":[2,2]}}

Ready to customize

Fine-tune Inkling for your domain on Tinker, our training platform.

Try on Tinker

Rightsholders and their authorized representatives can contact us about these models at [email protected].