Tether Launches 800 Million Parameter BotPsy Model, Tool Invocation Performance Surpasses 2.5 Times Larger Models

Tether launched the open-weight model QVAC BotPsy on November 11, featuring 800 million parameters and based on Alibaba's open-source Qwen3.5 architecture. This model can convert daily requests into local tool calls on users' own devices. Tether claims that in its internal testing, BotPsy set a new high in tool invocation evaluations among models with fewer than 1 billion parameters, surpassing models with 1.2 billion parameters and below. In four major evaluations—BFCL v4, ACEBench, τ-bench, and ToolSandbox—BotPsy's performance was approximately double that of the base model Qwen3.5-0.8B-Think, while also exceeding the performance of the Qwen3.5-2B-Think model, which has 2.5 times the parameters. Notably, BotPsy scored 46.1 points in the BFCL multi-turn tool invocation, nearly double the 23.8 points of the 2B model. BotPsy can reduce the need to send personal information such as emails and calendars to centralized service providers and can determine when no suitable tools are available, thus avoiding fabricated calls. Tether CEO Paolo Ardoino stated that users should decide what content the assistant can access and which external services it can connect to. The model weights have been published on Hugging Face.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.