meituan-longcat/LongCat-Flash-Omni

Omni-modal model with 560 billion parameters

View on index · View in 3D Map

Related repos

• Generates minutes-long videos from text and images
• Omnimodal model using mixture of experts
• audio-driven avatar video generation with prompt-based control and full-body animation
• Generates 12 billion avatars from strings
• fast attention algorithm for transformers, boosts speed and memory efficiency
• Perplexity clone using Gemini 2.0 Flash
• Run large language models locally, manage and interact with different model types
• multimodal AI model for real-time text, image, audio, and video chat
• API for OpenAI Whisper with diarization
• TypeScript SDK auto-selects best AI models