# Rimp

> All the AI models. One subscription. The unified API for image, video, voice, music, avatar, chat, 3D and upscale generation.

## Docs

- [Aggregate PAYG usage](https://docs.rimp.io/api-reference/aggregate-payg-usage.md)
- [Cancel a queued or running generation](https://docs.rimp.io/api-reference/cancel-a-queued-or-running-generation.md)
- [Chat completion (synchronous)](https://docs.rimp.io/api-reference/chat-completion-synchronous.md): Synchronous LLM chat completion across GPT-5, Claude, Gemini, Llama, DeepSeek, Mistral, Qwen and Grok. Credits are charged per output token (reconciled on completion).
- [Create a presigned upload URL for input assets](https://docs.rimp.io/api-reference/create-a-presigned-upload-url-for-input-assets.md)
- [Create a webhook](https://docs.rimp.io/api-reference/create-a-webhook.md)
- [Delete a webhook](https://docs.rimp.io/api-reference/delete-a-webhook.md)
- [Generate a 3D model (async)](https://docs.rimp.io/api-reference/generate-a-3d-model-async.md): Text-to-3D and image-to-3D via TRELLIS, Hunyuan 3D and Rodin. Output is a GLB mesh URL. Some models (Rodin) produce rigged characters.
- [Generate a talking avatar video](https://docs.rimp.io/api-reference/generate-a-talking-avatar-video.md): Lipsync + avatar generation via HeyGen and Sync Labs.
- [Generate an image (synchronous)](https://docs.rimp.io/api-reference/generate-an-image-synchronous.md)
- [Generate music from text prompt](https://docs.rimp.io/api-reference/generate-music-from-text-prompt.md): Full-length songs with vocals via MiniMax Music and Google Lyria.
- [Get a comparison with all child generations](https://docs.rimp.io/api-reference/get-a-comparison-with-all-child-generations.md)
- [Get a generation](https://docs.rimp.io/api-reference/get-a-generation.md)
- [List available models](https://docs.rimp.io/api-reference/list-available-models.md): Returns the live catalog of available models across all modalities with current pricing. Optionally filter by modality.
- [List webhooks](https://docs.rimp.io/api-reference/list-webhooks.md)
- [Queue a video generation (async)](https://docs.rimp.io/api-reference/queue-a-video-generation-async.md)
- [Run a multi-model comparison (async)](https://docs.rimp.io/api-reference/run-a-multi-model-comparison-async.md): Send one prompt to N models and reserve credits for all of them atomically. If the wallet cannot cover the sum, nothing is charged and the call returns 402. The number of models is capped by your plan (Free 2, Pro 3, Studio 6, Team 12).
- [Synthesize speech from text (TTS)](https://docs.rimp.io/api-reference/synthesize-speech-from-text-tts.md): Generate spoken audio with Inworld, MiniMax, OpenAI TTS and other voice providers.
- [Upscale or restore an image (async)](https://docs.rimp.io/api-reference/upscale-or-restore-an-image-async.md): Image upscaling (2x / 4x) and face restoration via Real-ESRGAN, Clarity, Topaz, Recraft and GFPGAN.
- [Authentication](https://docs.rimp.io/authentication.md): Bearer tokens, scoping, rotation, and Enterprise IP allowlist.
- [Changelog](https://docs.rimp.io/changelog.md): API changes, new models, and SDK releases. Follows Keep a Changelog.
- [Idempotency](https://docs.rimp.io/concepts/idempotency.md): Safely retry POST requests without double-charging.
- [Multi-model](https://docs.rimp.io/concepts/multi-model.md): How Rimp routes prompts across 8 modalities and ~60 models, and how to compare them side-by-side in one call.
- [Rate limits](https://docs.rimp.io/concepts/rate-limits.md): Per-key RPM limits, headers, backoff strategy, and how to raise limits.
- [Webhooks](https://docs.rimp.io/concepts/webhooks.md): Receive `generation.completed` events instead of polling. HMAC-signed and retried for 24 hours.
- [Errors](https://docs.rimp.io/errors.md): Stripe-style error envelope, retry semantics, and full code reference.
- [Welcome to Rimp](https://docs.rimp.io/introduction.md): One API for every image and video generation model that matters.
- [Connect via MCP](https://docs.rimp.io/mcp.md): Point Claude, Cursor, and other AI tools at the Rimp docs through the Model Context Protocol.
- [Quickstart](https://docs.rimp.io/quickstart.md): Make your first image and video generation in under 5 minutes.
- [curl recipes](https://docs.rimp.io/sdks/curl.md): Hit the API directly with bash + curl. Useful for testing, debugging, CI scripts, and language ports.
- [Python SDK](https://docs.rimp.io/sdks/python.md): Official Python client for the Rimp API. Sync first, async-friendly via httpx.
- [TypeScript SDK](https://docs.rimp.io/sdks/typescript.md): Official TypeScript / JavaScript client for the Rimp API.

## OpenAPI Specs

- [openapi](https://docs.rimp.io/api-reference/openapi.json)