# Longcat Image > AI image generation and editing platform powered by Meituan's open-source LongCat-Image 6B model, with industry-leading Chinese-English text rendering and photorealistic output. ## Docs - [AI Image Generator](https://longcat-image.org/ai-image-generator): Browser-based generator running LongCat-Image plus FLUX 2, Seedream 4, Z-Image Turbo and Nano Banana Pro; includes an image-to-image editing tab using LongCat-Image-Edit. - [LongCat Image Edit](https://longcat-image.org/longcat-image-edit): Dedicated single-reference editor that opens in image-to-image mode with fal's LongCat-Image-Edit model; covers object edits, text replacement, style changes, and practical editing limits. - [Pricing](https://longcat-image.org/pricing): Credit-based pricing with free starter credits on signup; one-time credit packs, no subscription required. - [Showcases](https://longcat-image.org/showcases): Gallery of images generated with LongCat-Image, including bilingual in-image text examples. ## Blog - [How to Use LongCat-Image Online](https://longcat-image.org/blog/how-to-use-longcat-image-online): Step-by-step guide to generating with LongCat-Image in the browser, prompting tips for accurate Chinese and English in-image text, and all access routes (online, fal.ai API, local). - [Run LongCat-Image Locally: ComfyUI, VRAM, GGUF](https://longcat-image.org/blog/longcat-image-comfyui-local-guide): Local deployment guide with primary-source numbers — 29.3 GB full pipeline, ~17 GB VRAM with CPU offload per the official README, native ComfyUI support since March 2026, community GGUF files from 2.1 GB with explicit quality caveats. - [LongCat-Image vs Qwen-Image](https://longcat-image.org/blog/longcat-image-vs-qwen-image): Benchmark-sourced comparison — LongCat 90.7% vs Qwen ~57% on the 8,105-character ChineseWord benchmark (cross-confirmed by both technical reports), while Qwen wins on aesthetics, ecosystem and API price. - [LongCat-Image vs FLUX](https://longcat-image.org/blog/longcat-image-vs-flux): Version-specific comparison centered on FLUX.2 Klein 4B, with FLUX.1 Dev benchmark context and FLUX.2 Dev as the quality ceiling; covers Chinese text, VRAM, editing, licenses, ecosystem, and price. - [LongCat-Image vs Z-Image Turbo](https://longcat-image.org/blog/longcat-image-vs-z-image): Two 6B Apache-2.0 models compared by Chinese-text evidence, speed, VRAM, released editing weights, and API cost, with Seedream 4.5 separated as a closed commercial benchmark. - [What is Longcat Image](https://longcat-image.org/blog/what-is-longcat-image): Platform overview — capabilities, use cases (marketing creatives, e-commerce, social content), and how the LongCat-Image 6B model differs from other generators. ## Key Facts - Runs Meituan's LongCat-Image 6B model (weights released 2025-12-05, Apache-2.0 license) - Chinese text rendering: 90.7% accuracy on the 8,105-character ChineseWord benchmark (arXiv 2512.07584), vs 56.6% for Qwen-Image and 58.5% for Seedream 4.0 - Supports text-to-image generation and instruction-based image editing (single reference image) - Free starter credits on signup; no credit card required to try - Multi-model generator: LongCat-Image, FLUX 2, Seedream 4, Z-Image Turbo, Nano Banana Pro ## Contact - Website: https://longcat-image.org - Support: support@longcat-image.org