Code Pluginsource linked

NVIDIA NIM Providerv1.1.0

NVIDIA NIM API provider for OpenClaw — 120+ live models including LLM chat, vision, embeddings, and image generation. Single API key, OpenAI-compatible endpoints. Tools fetch real-time model list from api.nvidia.com.

nvidia-nim-provider·runtime nvidia-nim-provider·by @534422530

Community code plugin. Review compatibility and verification before install.

openclaw plugins install clawhub:nvidia-nim-provider

Latest release: v1.1.0Download zip

Capabilities

Compatibility

Built With Open Claw Version

2026.5.26

Plugin Api Range

>=2026.3.24-beta.2

Security Scan

VirusTotal

Benign

View report →

OpenClaw

Benign

high confidence

✓

Purpose & Capability

The package purpose is NVIDIA NIM model discovery, pricing, recommendations, and cost estimates; the code registers five matching OpenClaw tools and only uses the NVIDIA API key for requests to NVIDIA's model-list endpoint.

ℹ

Instruction Scope

The manifest and README disclose the NVIDIA API key requirement and live API model fetching, though they could be clearer that requests go to NVIDIA-hosted services and that the bearer key is sent in request headers.

ℹ

Install Mechanism

It is a community code-plugin npm-pack with no install scripts and no dependency materialization, but it does execute plugin code on startup to register tools.

✓

Credentials

The observed runtime behavior is proportionate: no filesystem reads or writes, shell execution, local credential scraping, broad indexing, or account mutation; network use is limited to NVIDIA API calls needed for the advertised functionality.

ℹ

Persistence & Privilege

The plugin requires a user-provided NVIDIA API key in OpenClaw configuration and is activated on startup, but it does not create its own persistence, background worker, elevated privilege path, or hidden credential store.

Scan Findings in Context

[SQP-2] expected: The transparency concern is valid user guidance, but the behavior is purpose-aligned: the artifacts disclose a required NVIDIA API key and live API model fetching, and the inspected code only uses the key to call NVIDIA model metadata endpoints.

Assessment

Before installing, treat this as a networked community code plugin: only use a NVIDIA API key you are comfortable giving to OpenClaw configuration, understand that model-list requests are sent to NVIDIA with that key, and verify the publisher/source if you require stronger provenance.

Verification

Tier

source linked

Scope

artifact only

Summary

Validated package structure and linked the release to source metadata.

Source

github.com/laosi/contract-review-skill

Commit

HEAD

Tag

master

Provenance

Scan status

clean

NVIDIA NIM Provider Plugin

Access 42+ NVIDIA NIM API models through OpenClaw — LLM chat, vision analysis, text embeddings, and image generation. Single API key, OpenAI-compatible endpoints.

Features

List Models — Browse all available models by category
Model Info — Get detailed specs (context length, pricing, strengths)
Compare Pricing — Compare input/output costs across models
Recommend Model — AI-powered recommendations by use case and budget
Estimate Cost — Calculate API call costs before executing

Setup

Get an API key from build.nvidia.com
Configure in openclaw.json:

{
  "plugins": {
    "entries": {
      "nvidia-nim-provider": {
        "enabled": true,
        "config": {
          "apiKey": "nvapi-..."
        }
      }
    }
  }
}

Category	Description	Examples
llm	Chat completion	Nemotron, Llama, Qwen, GLM
vision	Image generation	SDXL Turbo, Playground
embedding	Text embeddings	NV-Embed-QA
rerank	Search reranking	NV-Rerank

Pricing

Pricing is in $/M tokens. Some models (Llama 3.1) are free. Check build.nvidia.com for latest pricing.