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).

All LawMatic product updates