Tether's BrainWhisperer project has achieved a 98.3% accuracy rate in converting brain signals into text, securing fourth place among 466 teams in the Brain-to-Text '25 Kaggle competition. The system, which utilizes OpenAI’s Whisper model with LoRA fine-tuning, achieved a word error rate of 1.78%. It decodes cortical electrical signals into text through a multi-model integration pipeline. In addition to this achievement, Tether is working on cross-individual signal decoding frameworks and developing non-invasive brain-computer interface (BCI) devices. The company has also launched Brain OS, an open-source brain operating system based on the QVAC platform, to further advance the field of brain signal processing.