NorthCUA-Nano
Computer-use agent - screenshot + task to GUI actions
None defined yet.
Computer-use agent - screenshot + task to GUI actions
Computer-use agent - screenshot + task to GUI actions
Class-conditional 512x512 pixel-space MoE diffusion
Prosody-aware speech LLM: speech in, text answer out
Streaming text-to-video with Wave Forcing on Wan2.1-T2V-1.3B
RL-refined 4-step Qwen-Image-2.1, T2I + editing
Speech + voice cloning with Audio8 0.6B TTS
Small English TTS with voice cloning, sounds and emotions
Real/Fake forensics for images and videos
Typed questions in, calibrated probabilities out
Rewrite prompts into MiniMax-H3 Ref2VA format
Physics-guided any-to-any bokeh editing
Detect AI-generated vs. real photos with AIRealNet-2
Two English voices from a 1.4M-parameter TTS stack
Spoken language ID for 23 Indian languages
Interleaved visual-textual spatial reasoning with WM-VLM
Multi-codebook diffusion LLM text-to-image
Omni-modal dense video captioning with parallel decoding
Rewrite prompts into MiniMax-H3 Ref2VA format
Ultra-fast Korean speech recognition (on-device STT)
Photo-oriented MOSS-Image finetune (text-to-image)
Texture 3D meshes from a prompt with SDXL projection
MOSS-Image 4-step + NVIDIA PiD 4x decode to ~4K
Visually-grounded multimodal reasoning with EVAN-8B