MiniMax H3-Image 上线 LMArena，支持 1600 万像素 10 秒生成
MiniMax 把新图像模型 H3-Image 放进了 LMArena Battle Mode，你能直接投票打分。它是开放权重的，4K 分辨率 10 秒出图，还支持 16 张参考图，做本地部署或想要高分辨率生图的可以试试。
MiniMax 的图像生成模型 H3-Image 现已进入 LMArena Battle Mode，用户可以提交创意提示词进行对比投票，官方评分即将公布。该模型支持 1600 万像素（16MP）4K 分辨率输出，单张生成时间约 10 秒。模型采用开放权重发布，已上传至 Hugging Face（MiniMaxAI/Mini…），同时可通过 MiniMax API、MiniMax Code 和 Hailuo AI 使用。
在能力层面，H3-Image 强调真实感画质、精确的提示词跟随与图像编辑，并且支持最多 16 张参考图输入，适合需要风格一致性或多主体控制的创作场景。上线周期间，16MP 及以上分辨率的生成享受五折优惠。
开放权重加上 4K 级分辨率和高参考图支持，让 H3-Image 与闭源图像生成服务形成差异化竞争，开发者和需要本地部署的团队可以直接从 Hugging Face 获取权重进行评估。
MiniMax H3-Image by @MiniMax_AI is now in the Arena! Find it in Battle Mode, bring your most creative prompts and start voting. Scores on the way. MiniMax (official) @MiniMax_AI 4K image generation in 10 seconds! ⚡ Meet MiniMax H3-Image. 16 million pixel resolution. Stunning realism. Precise prompt following & editing. Up to 16 reference images. All in 10 seconds. And it's open-weight. 🚀 Available now via our API, MiniMax Code, and Hailuo AI — with 50% off 16MP+ generations during launch week! Can't wait to see what you create. 🔥 Hugging Face huggingface.co/MiniMaxAI/Mini… g0 API platform.minimax.io/docs/guides/im… cM MiniMax Code: code.minimax.io/about CL
This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include MiniMax, H3-Image, LMArena, 图像生成.
Source attribution: curated from lmarena.ai (@lmarena_ai).
