Gradient™ AI 1-Click Models
Deploy popular AI models on DigitalOcean Gradient™ GPU Droplets with just a single click and start building absolutely anything your business needs.
1-Click Model Features
Deploy the most popular third-party models with just one click.
Zero Configuration
Reduce deployment time from weeks to minutes with zero-configuration setup, automatically optimizing the model and serving stack for your GPUs.
Optimized Deployments
Leverage maximum hardware efficiency, automatically. Experience high throughput, and minimal latency.
In YOUR Infrastructure
Deploy third-party AI models within your own environment. You’re in control of your droplet environment.
Plug and Play
Preconfigured, production-ready models from third-party providers. Skip the setup and start building with a single click.
Deploy DeepSeek LLM Models Instantly on Gradient AI GPU Droplets
DeepSeek R1 is now available as a 1-Click Model on Gradient AI GPU Droplets, enabling developers to launch and run powerful LLMs directly from the cloud console without manual setup.
1-Click Models
Optimized to run on high-performance GPU Droplets.
ERNIE 4.5-21B-A3B-Base
meta-llama/Meta-Llama-3.2-90B-Vision-Instruct
meta-llama/Meta-Llama-3.2-11B-Vision-Instruct
meta-llama/Meta-Llama-3.1-8B-Instruct
meta-llama/Meta-Llama-3.1-70B-Instruct
meta-llama/Meta-Llama-3.1-405B-Instruct-FP8
Qwen/Qwen2.5-7B-Instruct
google/gemma-2-9b-it
google/gemma-2-27b-it
mistralai/Mixtral-8x7B-Instruct-v0.1
mistralai/Mistral-7B-Instruct-v0.3
mistralai/Mixtral-8x22B-Instruct-v0.1
NousResearch/Hermes-3-Llama-3.1-8B
NousResearch/Hermes-3-Llama-3.1-70B
NousResearch/Hermes-3-Llama-3.1-405B
NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
DeepSeek/R1/671B
DeepSeekR1/Distill-Llama-8B
DeepSeek/R1/Distill-Qwen-7B
GPU Droplets starting at $0.76/GPU/hour
Experience affordable, high-performance AI infrastructure designed for developers and startups with DigitalOcean GPU Droplets.
Frequently asked questions about DigitalOcean 1-Click Models
What are Gradient 1-Click Models on DigitalOcean?
Gradient 1-Click Models let you instantly deploy third-party AI models—such as Hugging Face’s LLMs—on GPU Droplets with a single click. They require zero configuration, are optimized for GPU performance, and run on your own DigitalOcean infrastructure for fast, reliable AI inference.
How does zero-configuration deployment speed up AI model deployment?
With zero-configuration, the serving stack and AI model are automatically installed and optimized for GPU efficiency. This reduces deployment time from weeks to minutes, while ensuring high throughput and low latency for AI inference workloads.
Which third-party models are available via Gradient 1-Click Models?
Available models include Meta LLaMA 3, Mistral, Qwen, Gemma, Nous Hermes, and other popular Hugging Face models. All can be deployed instantly on GPU Droplets.
What are the main benefits of using GPU Droplets for AI model inference?
GPU Droplets provide hardware-accelerated performance designed for AI and machine learning tasks. They offer high throughput, low latency, and scalable deployments using NVIDIA-powered GPU infrastructure.
Can I deploy models within my own DigitalOcean infrastructure?
Yes. Gradient 1-Click Models are deployed on your own GPU Droplet environment, giving you full control over configuration, networking, and data security.
How do I interact with a deployed 1-Click Model?
Once deployed, you can connect to your GPU Droplet via SSH, retrieve the Bearer Token, and send inference requests—such as through cURL or Python—to the model’s endpoint running on port 8080.
Is it easy to deploy advanced LLM projects like a personal assistant?
Yes. DigitalOcean provides tutorials for creating advanced applications like voice-enabled personal assistants using tools like Gradio and FastAPI, powered by deployed 1-Click Models.
What differences exist between GPU Droplets and Bare Metal GPUs on Gradient?
GPU Droplets are virtualized GPU-powered VMs that are ideal for quick AI inference deployment. Bare Metal GPUs are dedicated single-tenant servers with multiple GPUs, suited for heavy training workloads or multi-node clustering.
Which AI model frameworks or providers are supported in Gradient 1-Click Models?
Gradient 1-Click Models integrate with Hugging Face’s model library, supporting a wide range of open-source LLMs optimized for GPU inference.
Why is “plug-and-play” deployment valuable for developers and ML teams?
Plug-and-play deployment means models are production-ready out of the box. Developers can skip time-consuming environment setup and focus on building applications, lowering both infrastructure complexity and time to market.