> ## Documentation Index
> Fetch the complete documentation index at: https://ai-docs.atlasflux.my/llms.txt
> Use this file to discover all available pages before exploring further.

# Model Catalog

> Overview of popular chat, image, and video models.

# Model Catalog & Capabilities

AtlasFlux AI now includes **over 400 models** across chat, image, video, music, and more. Below is an overview of popular categories. For the full list, visit the chat model selector.

## Chat Models

| Model                        | Best for                                                                          |
| ---------------------------- | --------------------------------------------------------------------------------- |
| GPT-5.5 Pro / GPT-5.4 Pro    | Complex reasoning, long-form writing, 1M context, token-based pricing.            |
| Claude Opus 4.8 / 4.7        | Advanced multimodal analysis, nuanced instructions, safety-critical applications. |
| Gemini 3.1 Pro               | Search integration, code understanding, broad knowledge.                          |
| DeepSeek R1 / V3.2           | Reasoning-exposed models, chain-of-thought, economical.                           |
| Grok 4.20 / 4.20 Multi-Agent | Real-time data (X integration), agentic tool calling.                             |
| Qwen 3 Max / Qwen 3.5        | Vision-capable, multilingual, free preview available.                             |
| Mistral Large                | Privacy-focused, strong European language performance.                            |
| Perplexity Sonar Pro Search  | Integrated search engine, perfect for fact-checking.                              |

## Image Models

| Model                       | Best for                                                          |
| --------------------------- | ----------------------------------------------------------------- |
| FLUX.2 Pro / Max            | Photorealistic generation, aspect ratio, img2img.                 |
| Imagen 4 / Ultra            | Google's flagship, exceptional prompt adherence, up to 4K.        |
| Seedream 4.5                | ByteDance model with multi-step reasoning, sequential generation. |
| Nano Banana Pro             | Up to 14 reference images, multiple aspect ratios.                |
| Ideogram V3 / Recraft V4    | Graphic design, text rendering, branding.                         |
| Grok Imagine                | High-realism, text-to-image and editing.                          |
| Proteus V0.3 / Lucid Origin | Community-tuned photorealism and anime.                           |

## Video Models

| Model                    | Best for                                         |
| ------------------------ | ------------------------------------------------ |
| Kling 2.6 / V3 Omni      | Cinematic quality, 4K, audio support.            |
| Veo 3.1 / Fast           | Google's premium model, JSON prompt support.     |
| Seedance 2.0 / 1 Pro     | Multi-image reference, complex motion.           |
| Grok Imagine Video       | xAI's text-to-video with native audio.           |
| PixVerse V4 / Transition | Fast social media content, creative transitions. |