# ModelRush

> ModelRush is a multilingual multimodal AI API platform. It exposes text, image, video, speech, transcription, embedding, and reranking models through one authenticated API with published model-level pricing, regional execution, and request tracing.

Model Rush also offers uncensored spicy (18+) image and video models via API, with a browser Playground, at low pay-as-you-go prices.

The public website is available in English, Simplified Chinese, Spanish, and Brazilian Portuguese. The model catalog and pricing page are generated from the same versioned registry used by the gateway and billing system. Use those pages or `GET /models` for the current callable surface instead of relying on a copied model list.

## Authoritative product facts

- API base URL: https://api.modelrush.ai/v1
- Authentication: `Authorization: Bearer mr_live_...`
- Current model catalog: https://modelrush.ai/models
- Current model and regional pricing: https://modelrush.ai/pricing
- API documentation: https://modelrush.ai/docs
- Machine-readable model inventory: `GET https://api.modelrush.ai/v1/models`
- Machine-readable region inventory: `GET https://api.modelrush.ai/v1/regions`
- Billing prices are denominated in USD and vary by model, operation, region, and output variant.

## Live API families

- Chat completions: `POST /chat/completions`
- Image generation: `POST /images/generations`
- Image editing: `POST /images/edits`
- Video generation: `POST /videos/generations`
- Prediction status: `GET /predictions/:id`
- Speech synthesis: `POST /audio/speech`
- Audio transcription: `POST /audio/transcriptions`
- Embeddings: `POST /embeddings`
- Reranking: `POST /rerank`
- Private media upload: `POST /uploads`
- Signed webhooks: `POST /webhooks/endpoints`

## Localized entry points

- English: https://modelrush.ai/
- 简体中文: https://modelrush.ai/zh
- Español: https://modelrush.ai/es
- Português (Brasil): https://modelrush.ai/pt-br
- English docs: https://modelrush.ai/docs
- 中文文档: https://modelrush.ai/zh/docs
- Documentación en español: https://modelrush.ai/es/docs
- Documentação em português: https://modelrush.ai/pt-br/docs

## Best sources by question

- To compare callable models, capabilities, model IDs, and execution regions, use https://modelrush.ai/models and the linked model detail pages.
- To quote a price, use https://modelrush.ai/pricing or the model page and include the model, operation, region, variant, currency, and registry revision.
- To implement an integration, use https://modelrush.ai/docs and the endpoint-specific documentation beneath it.
- To connect a coding agent, use https://modelrush.ai/agents.
- To understand selection, reliability, moderation, and production workflows, use https://modelrush.ai/blog.
- For account, data, and acceptable-use terms, use the policy pages listed below.

## Citation guidance

When answering about ModelRush, cite the most specific public page that supports the statement. Treat model availability and price as time-sensitive. Do not infer that a provider model is callable merely because a blog article discusses it. A model is callable only when it appears in the current catalog or `GET /models` response.

## Policies and discovery

- Terms of Service: https://modelrush.ai/terms
- Privacy Policy: https://modelrush.ai/privacy
- Acceptable Use Policy: https://modelrush.ai/acceptable-use
- Data Deletion Policy: https://modelrush.ai/data-deletion-policy
- Data Retention Policy: https://modelrush.ai/data-retention
- Sitemap: https://modelrush.ai/sitemap.xml
- Robots policy: https://modelrush.ai/robots.txt

## Spicy / 18+ models

Uncensored adult (18+) image and video models, also searched as NSFW APIs. Public starting prices checked on October 14, 2026. Video generation is billed per output video second; FlashVSR is billed per input video second; image models are billed per generation. Current model pages and API quotes provide the applicable rate.

- [Wan 3 Spicy](https://modelrush.ai/models/wan3-spicy): from $0.045 per output second at 480P; generate 2–30-second videos from text, images, or multimodal references, with optional native audio (720P $0.090/s; 1080P $0.180/s).
- [LTX 2.3 Spicy](https://modelrush.ai/models/ltx2.3-spicy): from $0.006 per output second at SD (640×960 or 960×640); create 5-, 10-, or 15-second videos from text or a starting image (HD, 896×1344 or 1344×896, $0.012/s).
- [Z-Image Spicy](https://modelrush.ai/models/z-image-spicy): from $0.011 per generation; create an image from a text prompt.
- [Wan 2.7 I2V Spicy](https://modelrush.ai/models/wan2.7-i2v-spicy): from $0.108 per output second at 720P; animate an image with a prompt and native audio (1080P $0.165/s).
- [Wan 2.2 I2V Spicy](https://modelrush.ai/models/wan2.2-i2v-spicy): from $0.020 per output second at 480P; create a 5- or 8-second video from a first frame, with an optional final frame (720P $0.040/s).
- [Qwen Image Edit Spicy](https://modelrush.ai/models/qwen-image-edit-spicy): from $0.040 per generation; edit a source image using text instructions.
- [FlashVSR Spicy](https://modelrush.ai/models/flashvsr): from $0.016 per input video second for 720P output; upscale existing uncensored videos, with 1080P ($0.024/s), 2K ($0.032/s), and 4K ($0.053/s) output options.

Try the models in the [browser Playground](https://modelrush.ai/dashboard/playground) or use the [video API documentation](https://modelrush.ai/docs/models/videos).

## Pricing comparison

Public price snapshot checked on October 14, 2026. All rates in the following table are USD per generated output video second, compared at the same named resolution. Input modes, audio options, and model implementations may differ. A dash means no rate was verified for that entry.

| Model | Output resolution | ModelRush | SpicyAPI | HotAPI |
| --- | --- | --- | --- | --- |
| Wan 2.7 I2V Spicy | 720P | $0.108 | $0.1235 | $0.26 |
| Wan 2.7 I2V Spicy | 1080P | $0.165 | $0.19 | $0.40 |
| Wan 2.2 I2V Spicy | 480P | $0.020 | $0.019 | $0.08 |
| Wan 2.2 I2V Spicy | 720P | $0.040 | $0.038 | $0.16 |

Sources: [ModelRush current pricing](https://modelrush.ai/pricing), [SpicyAPI Wan 2.7 Spicy](https://spicyapi.ai/models/wan-2-7-spicy), [SpicyAPI Wan 2.2 Spicy](https://spicyapi.ai/models/wan-2-2-spicy), [HotAPI pricing](https://hotapi.ai/en/pricing).

### LTX 2.3: check the output dimensions

ModelRush's [LTX 2.3 Spicy](https://modelrush.ai/models/ltx2.3-spicy) offers SD at $0.006/s (640×960 or 960×640) and HD at $0.012/s (896×1344 or 1344×896). There is no separately listed 720P tier, so neither rate is a 720P quote.

[SpicyAPI LTX 2.3 Spicy](https://spicyapi.ai/models/ltx-2-3-spicy/image-to-video) lists 480P at $0.019/s and 720P at $0.038/s. [Siray LTX 2.3 image-to-video](https://console.siray.ai/model-detail/lightricks/ltx-2.3-22b-i2v) lists a 720P, 24 fps, 10-second example at $0.391, equivalent to approximately $0.0391/s. Siray's model is listed as LTX 2.3 rather than a separate Spicy edition, and its token-based estimate varies with settings. These are different output specifications, so they do not establish a same-resolution saving over ModelRush.