Tether Releases 0.8B On-Device Function Calling Model BotPsy
Understand the story
On November 11, 2026, Tether AI Research announced on the Hugging Face Blog the release of the on-device personal assistant model BotPsy. The model has 0.8B parameters, targets function calling and agent workflow scenarios, and was trained based on Qwen3.5-0.8B. The weights will be released under the Apache 2.0 license, available in bf16, Q8, and Q4 precision versions, covering both safetensors and GGUF formats. The release has been confirmed via the official first-party channel; the subsequent open-source rollout remains to be seen.
AI overview of reporting · Updated 2 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- Hugging Face BlogBotPsy: On-Device Personal Assistant with a Sub-Billion-Parameter Agentic Language Model
Tether AI Research released BotPsy, an 0.8B-parameter on-device personal assistant model built for function calling and agent workflows, trained on Qwen3.5-0.8B. It will be released as Apache 2.0 open weights, offering safetensors and GGUF in bf16, Q8, and Q4.
Story heat trend
Current heat9·Comparable peak10（11/11 14:00）·Comparable change over 24 hours –
The trend compares the same continuously observed sources, which may cover fewer sources than the current heat score. Hover or select the chart to inspect each hour; use the left and right arrow keys to navigate.