LM Studio Setup: Run Your First Local LLM

Official requirements list Apple Silicon and macOS 14 or later for Mac; Intel Macs are not supported. Windows x64 systems require AVX2. LM Studio recommends 16GB or more RAM.

An 8GB Mac can test smaller models with modest context, but it should not be presented as a universal minimum.

For verification, save the model and runtime versions, source input, relevant settings, and observed output together. Repeat the step while changing only one factor, and record unexpected results and untested limits as carefully as successes before applying the guidance to private or production data.

Search for an instruct model and inspect the publisher, model card, license, parameter size, and GGUF quantization. LM Studio recommends 4-bit or higher when the machine can support it.

Start with one 3B–8B model rather than downloading many variants at once.

Preserve the before-and-after state and the time of the check so that another run can reproduce the result. Include at least one failure condition—such as empty input, constrained resources, or a restart—to reveal the boundary of the step rather than documenting only the happy path.

After download, select the model in Chat and load it into memory. Ask short, verifiable questions before increasing context or adding documents.

If loading fails, close other applications, reduce context, and try a smaller model or quantization.

Define completion with an observable result instead of a general impression. Repeat the same input, and if the output changes, isolate whether the model, runtime settings, or source data changed before moving to the next stage.

Official documentation states that downloaded models, chats, document conversations, and the local server can work offline. Model discovery, downloads, and updates need the network.

Test the exact workflow with connectivity disabled before relying on it for sensitive material.

For verification, save the model and runtime versions, source input, relevant settings, and observed output together. Repeat the step while changing only one factor, and record unexpected results and untested limits as carefully as successes before applying the guidance to private or production data.

The Developer view or `lms server start` can expose an API at localhost:1234. Keep the first application local to the same computer.

Configure authentication and network controls before binding to external interfaces.

Record the current version and settings before the example, then verify the expected response, file, or process afterward. Preserve the error and return to the smallest working command before adding options; this separates installation failures from input and integration failures.

lms server start

# Default local address
http://localhost:1234