SoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.
SoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.
A 4-GPU real-time version of SoulX-FlashTalk.
conda create -n flashtalk python=3.10
conda activate flashtalkpip install torch==2.7.1 torchvision==0.22.1 --index-url https://download.pytorch.org/whl/cu128pip install -r requirements.txtpip install ninja
pip install flash_attn==2.8.0.post2 --no-build-isolation# Ubuntu / Debian
apt-get install ffmpeg
# CentOS / RHEL
yum install ffmpeg ffmpeg-develor
# Conda (no root required)
conda install -c conda-forge ffmpeg==7| Model Component | Description | Link |
|---|---|---|
SoulX-FlashTalk-14B |
Our 14b model | Huggingface |
chinese-wav2vec2-base |
chinese-wav2vec2-base | Huggingface |
# If you are in china mainland, run this first: export HF_ENDPOINT=https://hf-mirror.com
pip install "huggingface_hub[cli]"
huggingface-cli download Soul-AILab/SoulX-FlashTalk-14B --local-dir ./models/SoulX-FlashTalk-14B
huggingface-cli download TencentGameMate/chinese-wav2vec2-base --local-dir ./models/chinese-wav2vec2-base# Infer on single GPU
# Requires more than 64G of VRAM. Use --cpu_offload to reduce VRAM usage to 40G.
bash inference_script_single_gpu.sh
# Infer on multy GPUs
# Real-time inference speed can only be supported on 8xH800 or higher graphics cards
bash inference_script_multi_gpu.shComing Soon!
If you are interested in leaving a message to our work, feel free to email [email protected] or [email protected] or [email protected] or [email protected] or [email protected]
Due to Group 1 reaching its capacity, we have opened a new WeChat group. Additionally, we represent SoulApp and warmly welcome everyone to download the app and join our Soul group for further technical discussions and updates!
If you find our work useful in your research, please consider citing:
@misc{shen2025soulxflashtalk,
title = {{SoulX-FlashTalk}: Real-Time Infinite Streaming of Audio-Driven Avatars via Self-Correcting Bidirectional Distillation},
author = {Shen, Le and Qiao, Qian and Yu, Tan and Zhou, Ke and Yu, Tianhang and Zhan, Yu and Wang, Zhenjie and Tao, Ming and Yin, Shunshun and Liu, Siyuan},
year = {2025},
eprint = {2512.23379},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
doi = {10.48550/arXiv.2512.23379},
url = {https://arxiv.org/abs/2512.23379}
}[!TIP] If you find our work useful, please also consider starring the original repositories of these foundational methods.
No open issues yet, or sync has not completed.