Pine AI getting 75.4% (SoTA) on τ³-Voice Leaderboard

Pine AI's model achieved a 75.4% score on the τ³-Voice Leaderboard, setting a new state-of-the-art (SoTA) for text agents resolving banking customer-service tasks over a ~700-document knowledge base, as published in τ-knowledge. The result was announced by Pine AI, which claims the top rank, surpassing previous models including Qwen 3.8 Max.

Text agents resolving banking customer-service tasks over a ~700-document knowledge base. Published as τ-knowledge. | Rank | Model | Released | Retrieval | Reasoning | User Sim | || |---|---|---|---|---|---|---|---| | 1 | Qwen 3.8 Max | | model release.release date fall back to the evaluation date.Have new results to share? Submit your model evaluation results by creating a pull request to add your JSON submission file. See our submission guidelines for the required format and process.