A powerful automation tool for generating one-click interactive chat and management scripts for NVIDIA NIM formatted large language models.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install nim-interactive-chat
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install nim-interactive-chat using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is NIM Interactive Chat?
NIM Interactive Chat is a developer-centric tool designed to simplify the deployment and interaction with NVIDIA NIM formatted models. By using Openclaw Skills, this tool automates the creation of a full suite of management scripts, including startup, shutdown, and an interactive CLI client. It effectively removes the friction of manual Docker management and health monitoring, providing a smooth experience similar to local chat applications while maintaining professional-grade inference capabilities.
NIM Interactive Chat Use Cases
- Rapidly deploying NVIDIA NIM format models locally for testing or development.
- Automating the container health check process to ensure a chat interface only opens when the model is ready.
- Simplifying the management of GPU memory utilization for NVIDIA inference microservices.
- Providing a consistent OpenAI-compatible API interface for local models using Openclaw Skills.
How NIM Interactive Chat Works
- The user executes the creation command providing the Docker image, container name, port, and model details.
- The skill generates three essential files: a startup script, a stop script, and a Python-based interactive chat client.
- Upon running the startup script, it automatically cleans up existing container conflicts and initiates the NIM Docker container.
- The script polls the service health check endpoint until the model is fully loaded into GPU memory.
- Once the service is healthy, the Python client is automatically launched to start a streaming conversation session.
NIM Interactive Chat Setup
To get started with this skill, ensure your environment has Docker, Python 3, and the openai Python package installed. You can generate your interactive environment using the following command structure:
create-nim-interactive <image-name> <container-name> <port> <model-name> [gpu-memory-utilization]
Example for deploying a Qwen model:
create-nim-interactive \
tgcr.turing-agi.com/public/qwen/qwen3.5-4b-spark-nim:latest \
qwen3.5-4b-nim \
18000 \
Qwen/Qwen3.5-4B \
0.88
NIM Interactive Chat Data Schema & Taxonomy
The skill generates a specific set of files to manage the model's lifecycle and interaction. This structure ensures that Openclaw Skills can maintain a clean and predictable workspace:
| File Type | Path Template | Description |
|---|---|---|
| Startup Script | ~/start-<container-name>.sh |
Handles container launch and automated health monitoring. |
| Stop Script | ~/stop-<container-name>.sh |
Safely stops and removes the specific model container. |
| Chat Client | ~/<container-name>-chat.py |
An OpenAI-compatible Python client for interactive streaming dialogue. |
NIM Interactive Chat Advanced Features
- Intelligent container lifecycle management that prevents port and naming conflicts automatically.
- Built-in wait logic that ensures the interaction only begins once the NVIDIA NIM service is fully operational.
- Customizable GPU memory utilization parameters to balance performance across different hardware configurations.
- Full support for streaming responses and conversation context retention within the generated Python client using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Wiki Maintainer
A professional standard operating procedure for AI agents to maintain, ingest, and query structured knowledge bases.
Minimax Tavily Search
A high-performance web search and content extraction skill powered by the Tavily API, specifically engineered for AI agents to retrieve noise-free, relevant data.
File Sorter by Extension
Automatically organizes files in a directory into extension-based folders to maintain clean and structured file systems.
File Organizer by Extension
A lightweight CLI tool that automatically sorts files into subdirectories based on their file extensions.
Company Profile: Debao Hengsheng (Medical SPD)
A professional profile skill providing comprehensive data on Debao Hengsheng's medical supply chain SPD services and smart hospital logistics.
SYSU Duck AI Companion
A specialized AI companion for Sun Yat-sen University students featuring local memory, gacha-style personalities, and campus-specific knowledge retrieval.
Featured*
