Model Catalog
50+ AI Models Behind One OpenAI-Compatible API
Every model below is reachable through the same Base URL, the same API key, and the same billing account. Filter by modality to shortlist candidates, then confirm cost, quota, and fallback routing before you integrate.
Text / Image / VideoOne API contract99.9% Uptime SLA5min Migration
Full Catalog
Model names are written exactly as they appear across this site. The catalog lists capability, not price: token-, request-, and duration-based billing units are explained on the pricing page so a model swap never hides a cost change.
Text Models
| Model | Vendor | Modality | Where it fits |
|---|---|---|---|
| gpt-5.6-sol (GPT-5.6 Sol) | OpenAI | Text Models | High-value complex work: deep reasoning, long-form analysis, hard coding tasks. |
| gpt-5.6-terra | OpenAI | Text Models | Balanced production traffic where quality, latency, and cost all matter. |
| gpt-5.6-luna-official | OpenAI | Text Models | High-volume lightweight work: classification, extraction, short replies. |
| gpt-5.4-mini-official | OpenAI | Text Models | Everyday chat and batch jobs that stay on a predictable budget. |
| gpt-6-astra | OpenAI | Text Models | Frontier experiments and complex generation before promoting a route to main traffic. |
| glm-5.3-flash | Zhipu | Text Models | Chinese-first chat and low-latency assistants. |
| deepseek-v4-flash | DeepSeek | Text Models | Code assistance and reasoning on a cost-optimized route. |
| kimi-k3 | Moonshot | Text Models | Long-context reading: documents, filings, and research briefs. |
| gemini-2.5-flash-lite-official | Google (Gemini) | Text Models | Lightweight multimodal understanding and captioning. |
| Grok 4.5 | xAI | Text Models | Reasoning and coding tasks alongside the GPT and Claude routes. |
Image Models
| Model | Vendor | Modality | Where it fits |
|---|---|---|---|
| GPT-Image-2.5 (flare / sunburst) | OpenAI | Image Models | flare for everyday, high-concurrency and batch generation; sunburst for precise editing and high-quality creative work. |
| GPT Image 2 | OpenAI | Image Models | Text-to-image and image-to-image for product and campaign visuals. |
| Gemini-3.1-Flash-Image Official | Google (Gemini) | Image Models | Fast text-to-image and image-to-image drafts. |
| Gemini-3-Pro-Image Official | Google (Gemini) | Image Models | Higher-fidelity image generation for final assets. |
| FLUX (Black Forest Labs) | Black Forest Labs | Image Models | Style-driven generation and restyling. |
| Gemini family (multimodal) | Google (Gemini) | Image Models | Image understanding and video-to-text summaries inside the same API contract. |
Video Models
| Model | Vendor | Modality | Where it fits |
|---|---|---|---|
| Veo3.1-quality-official | Video Models | High-quality text-to-video and image-to-video generation. | |
| Sora2 Official | OpenAI | Video Models | Text-to-video and image-to-video for concept previews and campaign prototypes. |
| Kling-v3 | Kuaishou | Video Models | Short-form video, storyboard drafts, and ad creative. |
| Vidu | Vidu | Video Models | Image-to-video animation from reference frames. |
| MiniMax | MiniMax | Video Models | Video generation for social and commerce content. |
Need a model that is not listed? Ask support and we will confirm availability instead of guessing an endpoint.
Authorized PartnersOpenAI18 models Anthropic4 models Google34 models ByteDance10 models Alibaba5 models DeepSeek3 models
Official model access with synchronized capability and reliable routing.
More providersBlack Forest Labs (FLUX)KuaishouMiniMaxMoonshotViduxAIZhipu
Route by Task
Short capability blocks that map a task to the models above. Use them to build route pools, then pick the primary and fallback model for each pool.
Generate brand-new images from text prompts.
Recommended models:Gemini-3.1-Flash-Image Official / Gemini-3-Pro-Image Official / GPT Image 2
Transform or refine existing images with controlled edits.
Recommended models:Gemini-3.1-Flash-Image Official / GPT Image 2 / Gemini-3-Pro-Image Official
Generate short video clips directly from textual instructions.
Recommended models:Veo3.1-quality-official / Kling-v3 / Sora2 Official
Animate still images into motion video outputs.
Recommended models:Veo3.1-quality-official / Kling-v3 / Sora2 Official
Convert video content into transcript-like text and concise summaries.
Recommended models:Gemini family (multimodal)
Pick the model tier that matches quality, cost, and throughput.
Recommended models:GPT-5.6 Sol (flagship) / GPT-5.6 Terra (balanced) / GPT-5.6 Luna (high volume) / Grok 4.5
Next Steps
Confirm the billing unit and the default route on the pricing page, then follow the migration guide to move your Base URL and API key onto the gateway.
Ready to route 50+ models with one API?
Shortlist models here, confirm cost and default routing on the pricing page, then run the 5-minute migration from the docs.