BotPsy brings a 0.8B-parameter agent to on-device assistants
Tether AI Research has introduced BotPsy, a 0.8B-parameter language model built on Qwen3.5-0.8B for function calling and agentic workflows. The model is intended for on-device personal-assistant tasks including web research, travel, commerce, finance, device control, file management, calendars, and messaging. Its developers say it is best suited to low- and medium-difficulty tasks and is not designed for high-stakes use.
In tests reported by the developers, BotPsy outperformed every evaluated model with up to 1.2B parameters on six tool-use and agentic benchmarks, leading on five. It also exceeded Qwen3.5-2B-Think on five of six benchmarks, despite having fewer parameters. Compared with its base model, scores improved on tool use, instruction following, and selected knowledge evaluations. The reported unsafe-response rate in automated red-teaming was 21%, down from 36% for the base model.
BotPsy will be released with open weights under the Apache 2.0 license in bf16, Q8, and Q4 formats. The model is available as Hugging Face safetensors and GGUF files for the QVAC SDK and llama.cpp. The training process used supervised fine-tuning, multi-teacher on-policy distillation, and two reinforcement-learning stages.
Written by Sakhanda Wire AI from reporting by Hugging Face. Facts, quotes and full context are in the original article below.