LM Studio

Run local GGUF models with LM Studio's OpenAI-compatible server, directly or through Yunxin.

Overview

LM Studio is a desktop application for running GGUF models locally. It exposes an OpenAI-compatible server (default http://localhost:1234/v1) so you can call your local models with standard tooling.

Official Website: https://lmstudio.ai Documentation: https://lmstudio.ai/docs

Key features

Local inference — Run models entirely on your own machine
Desktop GUI — Browse, download, and manage models with a friendly interface
Any GGUF model — Load any model in the GGUF format
OpenAI-compatible API — Standard chat completions endpoint, no key required locally
GPU acceleration — CUDA, Metal, and ROCm support

Calling LM Studio directly (local)

When you run LM Studio locally, point the OpenAI SDK at its local server. No API key is required, so any non-empty placeholder works.

from openai import OpenAI

client = OpenAI(
    api_key="lm-studio",
    base_url="http://localhost:1234/v1"
)

response = client.chat.completions.create(
    model="model-id",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "lm-studio",
  baseURL: "http://localhost:1234/v1",
});

const response = await client.chat.completions.create({
  model: "model-id",
  messages: [{ role: "user", content: "Hello!" }],
});

console.log(response.choices[0].message.content);

curl http://localhost:1234/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "model-id",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Using LM Studio through Yunxin

If your LM Studio server is registered with Yunxin, address its models with the lmstudio/<model> prefix and use the standard Yunxin base URL (https://api.yuhuanstudio.com/v1) and your sk- API key. This lets you mix local models with hosted providers under one client.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["YUNXIN_API_KEY"],
    base_url="https://api.yuhuanstudio.com/v1"
)

response = client.chat.completions.create(
    model="lmstudio/model-id",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.YUNXIN_API_KEY,
  baseURL: "https://api.yuhuanstudio.com/v1",
});

const response = await client.chat.completions.create({
  model: "lmstudio/model-id",
  messages: [{ role: "user", content: "Hello!" }],
});

console.log(response.choices[0].message.content);

curl https://api.yuhuanstudio.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YUNXIN_API_KEY" \
  -d '{
    "model": "lmstudio/model-id",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Available models

LM Studio serves whichever GGUF models you have downloaded and loaded. When accessed through Yunxin, query the Models API:

curl "https://api.yuhuanstudio.com/v1/models?provider=lmstudio" \
  -H "Authorization: Bearer $YUNXIN_API_KEY"

Capabilities depend on the loaded model and your hardware. The authoritative list for Yunxin is GET /v1/models?provider=lmstudio; for the local server, the loaded models are listed at http://localhost:1234/v1/models.

Official resources

Overview

Official Website: https://lmstudio.ai Documentation: https://lmstudio.ai/docs

Key features

Local inference — Run models entirely on your own machine
Desktop GUI — Browse, download, and manage models with a friendly interface
Any GGUF model — Load any model in the GGUF format
OpenAI-compatible API — Standard chat completions endpoint, no key required locally
GPU acceleration — CUDA, Metal, and ROCm support

Calling LM Studio directly (local)

When you run LM Studio locally, point the OpenAI SDK at its local server. No API key is required, so any non-empty placeholder works.

from openai import OpenAI

client = OpenAI(
    api_key="lm-studio",
    base_url="http://localhost:1234/v1"
)

response = client.chat.completions.create(
    model="model-id",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "lm-studio",
  baseURL: "http://localhost:1234/v1",
});

const response = await client.chat.completions.create({
  model: "model-id",
  messages: [{ role: "user", content: "Hello!" }],
});

console.log(response.choices[0].message.content);

curl http://localhost:1234/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "model-id",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Using LM Studio through Yunxin

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["YUNXIN_API_KEY"],
    base_url="https://api.yuhuanstudio.com/v1"
)

response = client.chat.completions.create(
    model="lmstudio/model-id",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.YUNXIN_API_KEY,
  baseURL: "https://api.yuhuanstudio.com/v1",
});

const response = await client.chat.completions.create({
  model: "lmstudio/model-id",
  messages: [{ role: "user", content: "Hello!" }],
});

console.log(response.choices[0].message.content);

curl https://api.yuhuanstudio.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $YUNXIN_API_KEY" \
  -d '{
    "model": "lmstudio/model-id",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Available models

LM Studio serves whichever GGUF models you have downloaded and loaded. When accessed through Yunxin, query the Models API:

curl "https://api.yuhuanstudio.com/v1/models?provider=lmstudio" \
  -H "Authorization: Bearer $YUNXIN_API_KEY"

Overview

Key features

Calling LM Studio directly (local)

Using LM Studio through Yunxin

Available models

Official resources

On this page

LM Studio

Overview

Key features

Calling LM Studio directly (local)

Using LM Studio through Yunxin

Available models

Official resources

On this page