Discover and use NVIDIA NIM chat, embeddings, and model catalog APIs.
npx skills add https://github.com/rish2jain/paperresearchagent --skill nvidia-api
Please help me install this Agent Skill. Skill: nvidia-api Source: https://github.com/rish2jain/paperresearchagent/tree/main/.claude/skills/nvidia-api Command: npx skills add https://github.com/rish2jain/paperresearchagent --skill nvidia-api
This skill streamlines the process of discovering and using NVIDIA API platforms, enabling developers to quickly access NIM cloud endpoints, model catalog, and retrieval APIs for prototype and production AI applications.
Get an API key from the NVIDIA API Catalog, then try a simple chat completion against the NVIDIA API endpoint using your HTTP client. Authenticate and submit a basic user message to /v1/chat/completions, then experiment with different models from the catalog.
High-intent search queries and answers about installing and using this skill.
To use NVIDIA NIM for chat completions, authenticate with an API key from the NVIDIA API Catalog and send a user message to the /v1/chat/completions endpoint. You can then experiment with different models from the catalog using your HTTP client.
Yes, you can migrate from OpenAI-compatible endpoints to NVIDIA NIM for smoother transitions. This skill provides migration support and best practices to help you transition your AI applications to NVIDIA cloud endpoints efficiently.
NVIDIA provides NIM endpoints for both chat and embeddings, alongside dedicated retrieval APIs. These allow you to generate embeddings and access retrieval mechanisms for prototype and production AI applications.
Yes, you need an API key from the NVIDIA API Catalog to access cloud endpoints and navigate the model catalog. Once authenticated, you can submit requests to cloud-hosted models or use self-hosted NIM deployments.
Yes, NVIDIA NIM supports streaming for chat completions. You can authenticate against the NVIDIA API endpoint and use streaming to receive continuous responses for your prototype and production AI applications.
The best way to prototype with self-hosted NIM deployments is to follow quick-start workflows and best practices. Use the NVIDIA API Catalog for reference implementations and apply these patterns to your self-hosted models.
rish2jainDeploy and operate NVIDIA Inference Microservices for GPU-accelerated AI inference.
truchotIntegrate LLMs, RAG, and vector databases into web applications.
zhangzhang-111-iDesigns RAG pipelines with chunking, embedding, retrieval, and evaluation strategies.
copilot-ldExpose an HTTP LLM API client for OpenAI-compatible chat completions and embeddings.
