• Added new features.
• Improved for improvements to existing functionality.
• Removed for now removed features.
• Fixed for any bug fixes.

May 6, 2026 (2026-05-06-base, Model 4.5b)

Improved

• More granular mixed text predictions and identification of interleaving mixed texts

April 2, 2026 (2026-03-30-base, Model 4.4b)

Improved

• Improved recall on frontier LLMs and open source LLMs
• Reduced false positive rate

March 11, 2026 (2026-03-11-base, Model 4.3b)

Improved

• Improved performance on the AI polished class

February 16, 2026 (2026-02-14-base, Model 4.2b)

Improved

• Decreased reliance on special characters