NVIDIA NIM API
VerifiedOfficialby NVIDIA
Official listing: built and published by NVIDIA. More from NVIDIA
Serves open models as NVIDIA-hosted NIM endpoints through an OpenAI-compatible API, with chat, embedding, reranking and image models, for developers prototyping before self-hosting.
- Access
- API
- Auth
- API Key
- Skills
- 5
Skills
OpenAI-Compatible Chat
Runs chat completions on hosted open models at integrate.api.nvidia.com/v1, with SSE streaming and tool calling.
Embeddings and Reranking
Creates text and image embeddings and reranks passages with NeMo Retriever and other retrieval models.
Image Generation
Generates images with hosted visual models such as FLUX.
Safety Guard Models
Screens content safety, topic control and jailbreak attempts with NemoGuard models.
Self-Hosted NIM
Runs the same models as NIM microservices on your own cloud, data center or workstation.
Related Agents
OpenAI API
by OpenAI
Serves OpenAI models through the Responses and Chat Completions APIs, with function calling, file search, realtime…
Cohere
by Cohere
Provides Command models for chat, RAG, and tool use, plus Embed and Rerank models for search, via a REST API and…
Gemini API
by Google
Serves Google's Gemini models through the Gemini Developer API and Gen AI SDKs, with function calling, structured…
Meta Model API
by Meta
Serves Meta's Muse Spark, Muse Image, Muse Voice Transcribe, and SAM models through Responses, Chat Completions, and…