Model positioning
Gemini 2.5 Flash Preview TTS is Google's preview text-to-speech model, focused on low latency, cost-effectiveness, and controllable speech generation. It takes text as input and produces audio. (aistudio.google.com)
Capabilities and uses
- Can generate single-speaker or two-speaker speech, with voice style, accent, and speaking rate adjustable through instructions. (aistudio.google.com)
- Can be used for narration, audiobooks, podcasts, conversational content, and voice assistants, among other scenarios. (aistudio.google.com)
Limitations and version status
- This model is a preview and should not be confused with Live API models used for interactive audio. The current model documentation lists an input limit of 8,192 tokens and an output limit of 16,384 tokens; features such as Live API, function calling, context caching, and code execution are not supported. (aistudio.google.com)
- This entry corresponds to the version released in May 2025. On December 10, 2025, Google announced that an upgraded version would replace the May version; the documentation for the API ID currently bearing the same name lists its most recent update as December 2025, so the current API cannot be assumed to remain fixed to the May version. (blog.google)
Pricing
The following are the currently published Gemini Developer API prices for the model with the same name, in US dollars (USD), as checked on November 18, 2026; they cannot be assumed to be the historical prices for the original May 2025 version. (aistudio.google.com)
- Standard mode, paid tier: Text input costs 0.50 USD per 1 million tokens; audio output costs 10.00 USD per 1 million tokens. Input and output in the standard free tier are marked as free; specific quotas and applicable conditions are subject to the official page.
- Batch mode, paid tier: The pricing page lists text input at 0.25 USD per 1 million tokens and audio output at 5.00 USD per 1 million tokens; it is not available in the free tier. However, the model details page also states that Batch API is not supported. The official pages are inconsistent, and actual availability should be verified before use. (aistudio.google.com)
Release date and official sources
- Release date: May 20, 2025. The Google Gemini API changelog explicitly lists this API model and its single-speaker and two-speaker speech capabilities. (aistudio.google.com)
- Google I/O developer announcement
- Google model details
- Google Gemini API pricing
- Google December 2025 TTS update announcement