Flow - 4x faster than typing
Mini toolsLongCat models
Browse all models from this model family.
ID
Model
Company
Type
Primary task
Open source
-
By MeituanLongCat-2.0 is an open-source MoE language model with 1.6 trillion total parameters (~48B activated per token) and a 1M-token context window. It introduces LongCat Sparse Attention for efficient long-context processing and was pretrained on 35+ trillion tokens using AI ASIC hardware. Strong focus on coding, agentic, and reasoning tasks.NewTextReleased 14d ago
-
By MeituanLongCat-Video-Avatar-1.5 is Meituan LongCatโs open-source audio-driven avatar video model for single-character and multi-character human video generation.NewMultimodalReleased 1mo ago
-
By MeituanLongCat-AudioDiT-3.5B is Meituan LongCatโs diffusion-based text-to-speech model built directly in waveform latent space rather than mel-spectrogram space. It is designed for high-fidelity speech generation and zero-shot voice cloning, supports Chinese and English, and is positioned as a top-performing open model on the Seed benchmark for speaker similarity and intelligibility.AudioReleased 3mo ago
-
By MeituanLongCat Next is a multimodal LongCat model focused on compact yet capable visual and speech understanding. The official intro highlights strong performance despite a 28x compression ratio, with particular strength in text rendering, speech comprehension, low-latency voice conversation, and customizable voice cloning.MultimodalReleased 3mo ago
-
By Meituan560B parameter omni-modal MoE model (about 27B active) for real time audio-visual interaction, built on LongCat-Flash with multimodal perception and speech modules.MultimodalReleased 5mo ago
-
By MeiGen AILongCat-Video-Avatar is Meituanโs audio-driven avatar model built on LongCat-Video, generating super-realistic, lip-synced long videos from audio plus optional text and images, with stable identity, natural motion and support for multi-person scenes.VideoReleased 6mo ago
-
By Meituan6B parameter bilingual (Chinese-English) text-to-image foundation model focused on photorealism, strong Chinese text rendering and high quality image editing.ImageReleased 7mo ago
-
By MeituanLongCat-Video is a long-horizon text and image to video model that keeps identity, style, and motion consistent over extended clips. It supports in-place edits, camera and pacing control, and fast previews to delivery-quality renders.VideoReleased 8mo ago
-
By MeituanLarge reasoning model built on a 560B MoE backbone, activating about 18.6 to 31.3B parameters for advanced chain-of-thought, formal and agentic reasoning tasks.TextReleased 9mo ago
-
By MeituanChat-oriented LongCat-Flash variant, a 560B MoE language model with around 27B parameters active per token, tuned as a fast, non-thinking foundation for general and agentic tasks.TextReleased 10mo ago
