How to Deploy DeepSeek‑R1 on Alibaba Cloud Compute Nest in Minutes

Compute Nest Overview

Alibaba Cloud Compute Nest is a PaaS platform that enables service providers and users to deploy, deliver, and manage services on Alibaba Cloud.

DeepSeek‑R1 Overview

DeepSeek‑R1 is a high‑performance inference model released as a Lite preview on 20 Nov 2024 and officially on 20 Jan 2025. The model weights are open‑sourced under the MIT license. It achieves performance comparable to OpenAI’s models by applying reinforcement learning with a small amount of labeled data. An intelligent training arena generates dynamic problems and validates solutions in real time, improving mathematical, code, and reasoning capabilities.

Deploy DeepSeek‑R1 on Compute Nest

Open the Compute Nest service deployment page at https://computenest.console.aliyun.com/service/instance/create/cn-hangzhou?type=user&ServiceName=DeepSeek%20R1

Select payment type, instance type, and set the root password, then confirm the order.

Click “Create Now”.

Wait until the deployment status shows Completed and locate the instance in the list.

Switch to the resource view and log in to the underlying ECS instance (e.g., via Cloud Assistant).

Model Invocation

After logging into the ECS terminal, use the following curl commands to generate text or start a chat session with the DeepSeek‑R1 model (assuming the Ollama API is listening on port 11434).

## Text generation
curl http://localhost:11434/api/generate -d '{
  "model": "deepseek-r1:7b",
  "prompt": "Who are you?"
}'

## Chat
curl http://localhost:11434/api/chat -d '{
  "model": "deepseek-r1:7b",
  "messages": [
    { "role": "user", "content": "why is the sky blue?" }
  ]
}'