AI tools for Synthesis

This innovative online tool allows users to effortlessly transform static photographs into dynamic, romantic kissing videos using advanced synthesis technology. The process is d

ToolSpace is a research toolkit for turning tool-use traces into executable training and evaluation environments. It focuses on long-horizon agent tasks that require several sta

Audio8-TTS-Realistic-Air is a compact speech-generation model with multilingual synthesis and zero-shot voice cloning. Its 0.6-billion-parameter main model supports 23 languages

Ling-3.0-Flash-Sante is an open healthcare-specialized language model for medical reasoning and research workflows. Its hybrid-linear MoE architecture has 124B total and 5.1B ac

Kling 4.0 is an audio-visual generation model for creating realistic, directed video content. The hosted model combines visual motion with audio performance and is available wit

AstaBrief is Ai2's 8B open-weight model for producing cited scientific reports from questions and retrieved literature. It synthesizes supplied research excerpts into structured

AuK is a speech foundation model that generates and edits audio through natural-language instructions. It combines speech synthesis, spoken-content changes, enhancement, separat

FlaRe, or Floating Radiance Networks, is a research method for representing and rendering 3D scenes with a compact radiance field. The project focuses on floating radiance primi

IndexTTS 2.5 is a zero-shot neural text-to-speech foundation model for expressive multilingual voice synthesis. It builds on a transformer-based Text-to-Semantic module and a no

CrisperWhisper 2.0 is Nyra Labs' verbatim speech recognition model for transcribing how people actually speak. It supports controllable output styles, multilingual transcription

ID-V2V is an identity-preserving video restylization framework from Eyeline Labs for applying scene, lighting, and style changes to a source video while keeping human identity a

Wan-Dancer is a hierarchical music-to-dance video generation framework from Tongyi Lab for creating long, coherent dance videos from music and reference imagery. It is designed

Fish Audio S2.1 Pro is a production-grade neural text-to-speech model available through the Fish Audio API with the model string s2.1-pro-free. It is designed for developers who

QUEST is an open family of 2B to 35B deep research agents trained with fully synthetic rubric-tree tasks. The models target fact seeking, citation grounding, and report synthesi

Modality Forcing is a post-training recipe that turns a pretrained text-to-image model into a joint image-depth generator. It leverages the spatial priors inside T2I models to s

World Tracing is a generative pixel-aligned geometry method for predicting layered 3D worlds beyond the visible surface. From a single image or short dynamic input, it estimates

WavFlow is a raw-waveform audio generation model designed for high-fidelity audio synthesis without latent compression or VAE bottlenecks. It focuses on making this workflow mor

PanoWorld is a generative spatial world model designed for consistent whole-house panorama synthesis for VR-style tours from floorplans and style references. It focuses on makin

MoCam is a video re-camera method for novel view synthesis under challenging camera trajectories and incomplete geometry. It lets users generate new camera movements such as lar

SwiftI2V is an efficient high-resolution image-to-video generation framework that turns a still image into a temporally coherent video at up to 2K resolution. It targets a major

Khala is a music generation research project focused on scaling acoustic token language models toward high-fidelity song creation. It generates full musical examples from prompt