adaption

Building extremely efficient intelligence that evolves with our world.

[adaption @adaption_ai]

Work by @wooldridgemike, @PeterAsaro, @lucie_nlp, @IGurevych, @sarahookr, @ginasue, Peter Norvig, Raymond Perrault, @iyadrahwan, Samantha Shorey, @TobyWalsh, and @LukeZettlemoyer.

Read the report: https://t.co/7URUfeZkaN

[adaption @adaption_ai]

Adaption is proud to collaborate on @StanfordHAI’s One Hundred Year Study on Artificial Intelligence.

Released every 5 years, this year’s study examines how generative AI scales, AI infrastructure investment, advances in scientific discovery, and more.

[adaption @adaption_ai]

Own your intelligence: https://t.co/13EySvKgNs

[adaption @adaption_ai]

Value comes from AI that keeps learning inside the business, not from the biggest model.

@sarahookr at the @ftlive Future of AI summit with Ken Moore (@Mastercard), Philipp Heltewig (@cognigy), Nathan Marsh (@Bentley Systems), and @andrewtghill.

[adaption @adaption_ai]

Read the documentation: https://t.co/FhgQRxhFAy

[adaption @adaption_ai]

Adaption is now a plugin in @ChatGPT and Codex.

Describe the data you wish you had, and Adaption invents it. Already have data? Adaptive Data gets it ready for training, and AutoScientist adjusts the recipe to fit your custom model.

Better models start where your work already happens.

[adaption @adaption_ai]

Watch the full video: https://t.co/MFerogeUm1

[adaption @adaption_ai]

Can the Adaption team name a moment in computer science history from one tiny detail?

Co-founder @sudip_r0y and Scientific Lead @hugo_larochelle were put to the test with the team.

[adaption @adaption_ai]

@yuntiandeng @sarahookr @sudip_r0y Welcome Yuntian! Excited to have your expertise and research on reasoning and neural programs at Adaption.

[adaption @adaption_ai]

AutoScientist configured the full intelligence stack around the model, leaving its weights unchanged. Tested on Wisedocs’ Medical Long Context Reasoning benchmark.

Research by @sudip_r0y and the research team: https://t.co/cFV78McpJ3

[adaption @adaption_ai]

New Adaption research: with AutoScientist, a lower-cost model outperformed GPT-6 Astra and Gemini 4 Argon on long medical records, at up to 15.4x lower cost.

Correct answers jumped from 58% to 75%. No retraining.

[adaption @adaption_ai]

It also gave the most complete answers of any model tested. Completeness rose from 76% to 94%, ahead of Claude Opus 5.5 and Gemini 4 Argon.

[adaption @adaption_ai]

Watch the full video: https://t.co/YpcG2395IN

[adaption @adaption_ai]

The Adaption team rates AI’s biggest debates from 1 to 10.

@sarahookr, @sudip_r0y, @hugo_larochelle, and @weiyinko_ml gave open weights a unanimous 10. Parallel parking did not fare as well.

What’s your score for AI agents?

[adaption @adaption_ai]

The questions worth asking used to mean hours of analysis by hand. Now it takes one click.

Try Visualize: https://t.co/Swliemhdcq

[adaption @adaption_ai]

Last week, Adaption introduced Visualize. Ask a question about the data behind your AI. Get an interactive chart that shows what your model will learn.

Your data had a lot to say. See what the Adaption community has Visualized so far.

[adaption @adaption_ai]

Learn more: https://t.co/5ebTGcQnBb

[adaption @adaption_ai]

A model that knows your business and gets sharper with every task is the catalyst for faster innovation.

Co-founder @sarahookr will discuss turning AI into a competitive advantage at the @ftlive Future of AI conference with @Mastercard, @Cognigy, @BentleySystems.

[adaption @adaption_ai]

Learn more: https://t.co/b25WGHVfd8

[adaption @adaption_ai]

Millions of developers build in @cursor_ai. Now you can build custom models without leaving it.

Adaption is live on the Cursor Marketplace. Go from dataset to trained model, right from the Cursor agent.