Tether AI Research releases BotPsy, a 0.8B on-device agent model
AITether AI Research introduces BotPsy, a 0.8B-parameter model built on Qwen3.5-0.8B for function calling and agentic workflows on devices. The post reports BotPsy leads five of six tool-use benchmarks among models up to 1.2B parameters, and says it will be released with open weights under Apache 2.0 in bf16, Q8 and Q4 precisions, as safetensors and GGUF files for the QVAC SDK and llama.cpp.
AIWhy it matters.
The post compares BotPsy's tool-use benchmarks against several models at the same size and reports its weak spots, which helps judge fit for on-device agent work.
Source details
Hugging Face Blog · huggingface.coPublished · added here
Original title: BotPsy: On-Device Personal Assistant with a Sub-Billion-Parameter Agentic Language Model