LM Studio Blog

How to fine-tune FunctionGemma and run it locally

Step by step guide for fine-tuning FunctionGemma with Unsloth, and then running it in LM Studio

LM Studio 0.3.36

LM Studio 0.3.35

LM Studio 0.3.34

Ministral 3

LM Studio 0.3.33: Ministral 3 support, Olmo-3 tool calling, and release notes

LM Studio 0.3.32

GLM 4.5 tool calling, olmOCR-2, improved image input handling in `/v1/responses`, Flash Attention defaults for Vulkan/Metal, and bug fixes.

LM Studio 0.3.31

Image input improvements, MiniMax M2 tool calling, Flash Attention default for CUDA, new CLI runtime management, macOS 26 support, and bug fixes.

OpenAI gpt-oss-safeguard

Open safety reasoning models (120B and 20B) with bring-your-own-policy moderation, now supported in LM Studio on launch day.

NVIDIA DGX Spark

LM Studio now ships for Linux on ARM and launches with NVIDIA DGX Spark — a tiny but mighty Linux ARM box.

LM Studio 0.3.30

Bug fixes: Qwen tool-calling streaming, Vulkan iGPU loading, and `developer` role support in `/v1/responses`.

Use OpenAI's Responses API with local models

OpenAI-compatible `/v1/responses` endpoint (stateful chats, remote mcp, custom tools)

LM Studio 0.3.28

Default model‑variant selection in My Models, better RAM/VRAM estimates (mixed quantizations, mislabeled params, non‑transformers), and bug fixes.