Overview
Crusoe Cloud's managed inference service allows you to interact with supported models via our APIs. Pricing for each model is listed on model cards under the 'Inference' tab in the Crusoe Cloud console.
This capability is currently in Limited Availability, and may not be visible to your organization by default. To request access, please contact support.
Available Models
For text generation
We provide an OpenAI-API compatible endpoint at api.crusoe.ai for the models below. All Meta models provided by Crusoe are “Built with Llama”.
| Name | Provider | Type | Context Length | License | Acceptable Use Policy |
|---|---|---|---|---|---|
| meta-llama/Llama-3.3-70B-Instruct (Model card¹) | Meta | instruct | 128k | Llama 3.3 Community License Agreement | Llama 3.3 Acceptable Use Policy |
| openai/gpt-oss-120b (Model card¹) | OpenAI | instruct | 128k | Apache License 2.0 | Acceptable Use Policy |
| deepseek-ai/DeepSeek-V3-0324 (Model card¹) | DeepSeek | instruct | 160k | MIT License | MIT License |
| deepseek-ai/DeepSeek-R1-0528 (Model card¹) | DeepSeek | instruct | 160k | MIT License | MIT License |
| deepseek-ai/DeepSeek-V3.1 (Model card¹) | DeepSeek | instruct | 160k | MIT License | MIT License |
| Qwen3/Qwen3-30B-A3B (Model card¹) | Qwen | instruct | 40k | Apache License 2.0 | Apache License 2.0 |
| google/gemma-3-12b-it (Model card¹) | instruct | 128k | Gemma Terms of Use | Gemma Terms of Use, note use restrictions in Section 3.2 |
Video to Video Models
We've partnered with Decart, a leading AI lab focused on delivering real-time generative experiences to offer MirageLSD, the industry's first real-time video to video model through our managed inference service. We currently support a batch API allowing users to transform input videos with a prompt in near real-time, with a real-time streaming API coming soon. Usage of the model is covered under Crusoe's Platform Privacy policy and Acceptable Use Policy.
Key model details are outlined below. Details on how to interact with Video to Video models can be found in the Getting Started with Inference APIs section here.
| Category | Details |
|---|---|
Output Video Resolution | 1080p |
Output Video Framerate | 30 fps |
Output Aspect Ratio | 16:9 |
Recommended Input Video Resolution | >720p, 1080p is ideal |