MiniMax-H3-Turbo
4-step text/image-to-video with audio (MiniMax-H3)
None defined yet.
4-step text/image-to-video with audio (MiniMax-H3)
Typed questions in, a probability per option out
Causal chunk-by-chunk world-model video streaming
Board-conditioned LLM chess analysis and best move
Korean speech-to-text with TeamUNIVA's Qwen3-ASR-1.7B
Continuous two-sided fights with sound from MiniMax-H3
Few-step causal text-to-video with a 1.3B student
Typed decisions in, calibrated probabilities out
A decision model — calibrated option probabilities
Typed decisions with calibrated probabilities
Calibrated probabilities for typed decision questions
Typed decisions in, calibrated probabilities out
Speech in, clean intent-preserving text out
Typed, calibrated decisions from a state + question schema
6.82M-param text-to-image model, trained fully from scratch
Design a voice from a text description, then speak (zh)
Goal-directed video world model, step by step
Prompt + reference image to aesthetic HTML infographics
Prompt (+ reference image) to HTML/CSS infographic
Spatial-reasoning VLM distilled with routed 3D privilege
Multilingual zero-shot TTS and voice cloning
Visually grounded long-horizon information-seeking agent
Typed decisions with a probability for every option
Typed decisions with a probability for every option