TokenPost·
Tether’s 0.8B BotPsy Beats Larger Model in Five of Six AI Tests

The open-weight model averaged 45.0 across six tool-use evaluations and scored 86.1 for hallucination resistance on BFCL v4.
Every crypto headline. One feed. Updated every minute.

The open-weight model averaged 45.0 across six tool-use evaluations and scored 86.1 for hallucination resistance on BFCL v4.