LM Studio in AGNT
A desktop app for local models with an OpenAI-style server. Every AGNT workflow is provider-portable, so the same agents run here or anywhere else you point them.
Why teams pick it
Models
- any GGUF checkpoint
Strengths
- GUI model management
- drop-in local endpoint
- works offline
In AGNT
Agents, workflows and tools use LM Studio like any other provider — same nodes, same receipts, swap models per step if you want.
Local models with a graphical front door
LM Studio makes local inference approachable: browse and download checkpoints, manage what is loaded, and expose an OpenAI-compatible server that AGNT connects to like any other provider. For anyone who would rather not manage models from a terminal, it is the gentler path to the same destination.
What to watch
The OpenAI-compatible endpoint is convenient but not identical to the real thing in every edge case, so verify tool calling on the specific checkpoint you intend to use. As with any local setup, VRAM is the binding constraint on model choice.
Using LM Studio in a workflow
A workflow can use LM Studio for one step and another provider for the next. Teams typically reserve any GGUF checkpoint for work that rewards GUI model management, then hand bulk steps to a faster tier. The model behind an operation is a setting in AGNT, so this is tuned by measurement rather than decided up front.
Connect LM Studio in two minutes
- Download AGNT Community Core — free, local-first, no account needed to run.
- Start LM Studio on your machine, then pick it under Settings → Providers; AGNT finds the local endpoint.
- Give an agent a job — or install one from the marketplace — and watch the first receipt come back.
LM Studio in AGNT — common questions
How does AGNT connect to LM Studio?
Through its local OpenAI-compatible server — select it under providers and point at the local endpoint.
Can I use any GGUF model?
Yes, subject to what your hardware can hold.
Does it work offline?
Once a model is downloaded, inference runs with no network at all.