SafetyPage

Lab system cards and frameworks · Model card · Google DeepMind

Gemini 3.8 Audio (Live, Live Extended Thinking, Flash TTS, Flash-Lite TTS) Model Card

Sep 2026

Read at storage.googleapis.com ↗Read at sourceAll rights reserved

Outline · 7 pages

  1. Gemini 3.8 Audio (Live, Live Extended Thinking, Flash TTS, Flash-Lite TTS)p. 1
  2. Model cardp. 1
  3. Model Informationp. 2
  4. Description​​Gemini 3.8 Audio (Live, Live Extended Thinking, Flash TTS, Flash-Lite TTS) is an addition to the Gemini 3 series of highly-capable models. This model card describes the native capabilities (e.g., audio and video) as additional outputs of Gemini, including optional avatar capabilities (Live Avatar). Information specific to these modalities is specified in-line and referred to as Gemini 3.8 Live, Gemini 3.8 Live Extended Thinking, Gemini 3.8 Flash TTS, and Gemini 3.8 Flash-Lite TTS (collectively as Gemini 3.8 Audio).p. 2
  5. Model dependenciesp. 2
  6. Inputsp. 2
  7. Outputsp. 2
  8. Architecturep. 3
  9. Model Datap. 3
  10. Training Datasetp. 3
  11. Training Data Processingp. 3
  12. Implementation and Sustainabilityp. 3
  13. Hardwarep. 3
  14. Softwarep. 3
  15. Distributionp. 3
  16. Evaluationp. 5
  17. Approachp. 5
  18. Intended Usage and Limitationsp. 5
  19. Benefit and Intended Usagep. 5
  20. Known Limitationsp. 5
  21. Acceptable Usagep. 6
  22. Ethics and Content Safetyp. 6
  23. Evaluation Approachp. 6
  24. Safety Policiesp. 6
  25. Frontier Safety Assessmentp. 7
  26. Risks and Mitigationsp. 7