LawMaticAISettings and every LawMatic app now support the cloud provider NVIDIA NIM (NVIDIA API Catalog).
You can connect NVIDIA models through a single OpenAI-compatible API: one key from build.nvidia.com, configuration in the familiar “Cloud providers” section, a connection test, and model selection from the catalog — the same workflow as OpenRouter, Polza, and other aggregators.
What you get
- One key, many models — Qwen, Llama, Mistral, Nemotron, Kimi, and other NVIDIA catalog models without a separate integration per vendor.
- Cloud mode — runs via integrate.api.nvidia.com, with no Docker, GPU, or local NIM on your machine.
- Familiar UI — API key, base URL, and default model in the same settings as other LawMatic providers.
- Shared settings — configuration is stored in the ecosystem and can sync across LawMatic apps through the shared settings mechanism.
Who it is for
For developers and LawMatic app users who want to:
- try NVIDIA Catalog models in their scenarios (chat, agents, RAG);
- keep several providers in one place and switch between them;
- avoid running their own inference stack — cloud API only.
Getting started
- Get an API key at build.nvidia.com.
- In AI settings, choose the NVIDIA NIM provider.
- Paste the key and, if needed, run “Test and fetch models”.
- Pick a model from the list or from recommendations (for example, qwen/qwen3-next-80b-a3b-thinking).