Launch DeepSeek Models in Seconds with One‑Click Cloud Development

Quick Start

CNB includes pre‑built DeepSeek models (1.5B, 7B, 8B, 14B, 32B). To launch a model in seconds:

Fork the example repository: https://cnb.cool/examples/ecosystem/deepseek

Check out the branch that matches the desired model and open the Cloud Native Development environment.

When the remote development terminal appears (≈5‑10 s), run the following commands, replacing $ds with the model identifier (e.g., deepseek-r1:1.5b):

ollama serve &
ollama run $ds

These commands start the Ollama server and load the selected DeepSeek model for interactive chat.

Docker Alternative

If you prefer a containerized approach, execute:

docker run --rm -it docker.cnb.cool/examples/ecosystem/deepseek/1.5b:latest
ollama serve &
ollama run deepseek-r1:1.5b

The image contains the 1.5B model, starts Ollama, and runs the model without additional downloads or traffic fees.

Advanced Long‑Term Deployment

For production‑grade or long‑running deployments, use the Cloud Application Quick Deploy solution for DeepSeek. Detailed instructions are available at: https://mp.weixin.qq.com/s?__biz=MzA3NTE4NTgzMQ==∣=2648023588&idx=1&sn=fb7beb200c6d552157cf99432418a670&scene=21&nwr_flag=1#wechat_redirect