Hy4 is the official release of Tencent's Hy4 model, succeeding Hy4 Preview, with improvements in agentic capabilities and in long-horizon, real-world productivity tasks. It introduces visual understanding for the first time, accepting text and image input, and supports tool calling, structured outputs and a 1M-token co…
HY-Image 3.5 Preview is a unified image generation and editing model from Tencent Hunyuan, built on the Hunyuan Image 3.0 mixture-of-experts base. It handles text-to-image, image-to-image and multi-turn editing in one model, keeps subjects consistent across edits and renders Chinese and English text inside images.
Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that require planning, context continuity, and sustained multi-step execution.
Hy-MT2-1.8B is a compact 1.8B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.
Hy-MT2-30B-A3B is Tencent's flagship translation model in the Hy-MT2 family. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation. It uses 3B active parameters out of 30B total.
Hy-MT2-7B is a 7B-parameter translation model from Tencent. It supports 33 language pairs and five Chinese dialect and minority-language pairs, with workflows for structured, delimiter-based, contextual, glossary-based, and style-guided translation.
Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort: a direct no-think mode by default, plus low and high chain-of-thought modes for complex math, cod…
Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to balance speed and depth depending on the task, while delivering strong code generation and reliable per…
HY-WU is a scalable framework from Tencent Hunyuan for on-the-fly conditional generation of low-rank updates: it synthesizes instance-conditioned adapter weights from image and instruction and edits images without test-time optimization.
HunyuanImage-3.0 Instruct is a native multimodal model from Tencent that unifies multimodal understanding and generation within an autoregressive framework. The Instruct version adds reasoning for prompt enhancement and image-to-image generation, with performance comparable to or better than leading closed-source model…
Hunyuan Video 1.5 is Tencent's latest and best video model
Hunyuan Image 3.0 is Tencent’s native multimodal model for unified multimodal understanding and generation within an autoregressive framework. Its Mixture of Experts (MoE) architecture, the largest among open-source image generation models, has 80 billion parameters and 64 experts, and it produces photorealistic images…
FLUX.1 SRPO [dev] is Tencent's fine-tune of FLUX.1 [dev], a 12 billion parameter flow transformer, trained with Semantic Relative Preference Optimization for more realistic, aesthetic images from text. It is suitable for personal and commercial use.
HunyuanImage 2.1 is a two-stage text-to-image model from Tencent: a 17B-parameter diffusion transformer with a multimodal LLM and a character-aware text encoder for accurate multilingual text rendering, followed by a refiner for clarity.
Hunyuan-A13B is a 13B active parameter Mixture-of-Experts (MoE) language model developed by Tencent, with a total parameter count of 80B and support for reasoning via Chain-of-Thought. It offers competitive benchmark performance across mathematics, science, coding, and multi-turn reasoning tasks, while maintaining high…
