IndexTTS 2.5 needs about 9.3 GB of memory at its main file (a 7.9 GB file) with an 8K context window, so it fits on a 12 GB GPU.
IndexTTS 2.5 – zero‑shot text‑to‑speech with cross‑lingual voice cloning
IndexTTS 2.5 is an autoregressive zero‑shot TTS model that clones a voice from a single reference audio clip. It supports Chinese, English, Japanese, Spanish and Arabic, and offers speaking speed control and emotion control separate from timbre.
Good at
- Voice cloning from one clip
- Cross‑lingual transfer
- Emotion control via QwenEmotion
Limits
- Requires a reference WAV file
- No vision support
- Long text split into segments
Best for
- Generating synthetic speech for podcasts
- Creating multilingual voice assistants
- Producing voiceovers for videos
Who it’s for. Developers and researchers needing high‑quality, cross‑lingual TTS with controllable prosody. The model is suitable for applications that can provide a reference audio clip and handle segment‑based prosody limitations.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 7.9 GB | 9.3 GB | Too big | Get file → |
The original‑dtype GGUF file is 7.89 GB and needs approximately 9.3 GB of VRAM at 8 k context. If memory is limited, consider a lower‑precision quantization, but no quantized package is currently available in the file list.
What GPU do you need for IndexTTS 2.5?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is the main file (7.9 GB file, about 9.3 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about IndexTTS 2.5
How much VRAM does IndexTTS 2.5 need?
About 9.3 GB at its main file with an 8K context window. That is the 7.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run IndexTTS 2.5 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of IndexTTS 2.5 GGUF?
It depends on the quant: file is 7.9 GB.
Where can I download IndexTTS 2.5 GGUF?
The files are on Hugging Face in the Richasy/IndexTTS-2.5-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does IndexTTS 2.5 use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run IndexTTS 2.5 on 8 GB VRAM?
The original‑dtype model requires about 9.3 GB of VRAM at 8 k context, so 8 GB is insufficient without quantization.
Does IndexTTS 2.5 support emotion control?
Yes, but it requires the QwenEmotion model, which must be loaded with use_qwen_emo=True.
Is the model licensed for commercial use?
The model is released under the bilibili Model Use License Agreement; check the license terms before commercial deployment.