CCTest · Blog
LLaDA-Image: An Open Recipe for Unified Image Generation and Editing
Diffusion Models
cctest.ai
Diffusion Models

LLaDA-Image: An Open Recipe for Unified Image Generation and Editing

LLaDA-Image combines a from-scratch 6B Diffusion Transformer with a frozen vision-language understanding module for generation, editing, and text rendering. Its distilled Turbo variant reduces inference to 2–4 sampling steps while the project releases weights, code, and training recipes.

Read more
CCTest · Blog
Terminal-Universe Turns Agent Trajectories into Reusable Terminal Environments
AI Agents
cctest.ai
AI Agents

Terminal-Universe Turns Agent Trajectories into Reusable Terminal Environments

Terminal-Universe proposes reconstructing executable workspaces from existing terminal-agent trajectories instead of building every environment from scratch. The recovered environments can support original-task replay, new task synthesis, cross-repository queries, and multi-turn interactions.

Read more
CCTest · Blog
RoboTok Turns Internet Videos into a Searchable Data Engine for Dexterous Robots
Robotics & Physical AI
cctest.ai

RoboTok Turns Internet Videos into a Searchable Data Engine for Dexterous Robots

RoboTok retrieves human manipulation videos that resemble a given demonstration, using hand-motion similarity rather than relying mainly on visual appearance. Its actor-centered 3D hand-trajectory representation is designed to make retrieval more robust across viewpoints, occlusions, and changing scenes.

Read more