Skip to main content
Latest Update
2026-11-04 12:02 UTC

Cognition unveils SWE-2 Flash coding model built on Inkling-small

Cognition unveils SWE-2 Flash coding model built on Inkling-small

According to Soumith Chintala citing Cognition on X, Cognition introduced SWE-2 Flash, a fast coding model post-trained on Inkling-small that it says outperforms all American open-weight models on leading evaluations and will share weights securely with partners and researchers.

Source

Analysis

Cognition introduced SWE-2 Flash, its new fast and efficient coding model, on November 4, 2026. The model is post-trained on top of Inkling-small and outperforms all American open-weight models on leading evaluations, including those three times larger.

To balance transparency with cyber risks, Cognition is developing a program for secure weight sharing with partners and researchers, according to the company's announcement on X and commentary from Soumith Chintala.

Soumith Chintala

@soumithchintala

Cofounded and lead Pytorch at Meta. Also dabble in robotics at NYU.