Models API
The Runcrate Models API gives you access to 200+ AI models through a single endpoint. It’s fully OpenAI-compatible — use the same SDKs and code you already have, just change the base URL.Base URL
OpenAI SDK Compatibility
The API works with the official OpenAI Python and JavaScript SDKs:Endpoints
| Endpoint | Method | Categories | Description |
|---|---|---|---|
/v1/chat/completions | POST | Chat, Code, Reasoning, Vision | Text generation with streaming support |
/v1/images/generations | POST | Image | Text-to-image generation |
/v1/videos | POST | Video | Submit async video generation job |
/v1/videos/{id} | GET | Video | Poll video job status |
/v1/videos/{id}/download | GET | Video | Download completed video |
/v1/audio/speech | POST | TTS | Text-to-speech synthesis |
/v1/audio/transcriptions | POST | ASR | Speech-to-text transcription |
Model Categories
| Category | Models | Use Cases |
|---|---|---|
| Chat | DeepSeek-V3, GPT-4o, Claude 4, Llama 4, Qwen3, Gemini 2.5, Mistral | Conversation, instruction following, general tasks |
| Reasoning | DeepSeek-R1, QwQ, Claude extended thinking | Math, logic, complex problem-solving |
| Code | Codestral, DeepSeek-Coder, Qwen-Coder | Code generation, debugging, analysis |
| Vision | Gemini 2.5, Llama 4 Maverick, Gemma 3, GPT-4o | Image understanding, visual Q&A |
| Image | FLUX.1, FLUX.2, Stable Diffusion, Ideogram | Text-to-image, image editing |
| Video | Sora 2, Veo 3, Kling v3, Seedance, Hailuo | Text-to-video, cinematic generation |
| TTS | Kokoro, Orpheus, Qwen3-TTS, Zonos, Chatterbox | Voice synthesis, audiobook generation |
| ASR | Whisper | Audio transcription |
Pricing
Models are billed per token (text) or per generation (images, video, audio). Pricing varies by model. Check the Model Catalog or Pricing page for rates.Next Steps
Authentication
Get your API key and authenticate requests
Chat Completions
Start with the most popular endpoint