LM Studio
Desktop app for downloading and running AI models locally on macOS, Windows and Linux
In our rankings
Scores on the apps for running ai models locally criteria
Evaluated Nov 5, 2026- Model formats and downloads ×3 5
5/5 means: Two or more model formats, in-app Hugging Face search and download, and import of local model files
Runs GGUF and, on Apple Silicon, MLX models; searches and downloads from Hugging Face in the app; and can sideload model files from outside the app.
3 quotes - GPU support ×3 4
4/5 means: NVIDIA, AMD and Apple Silicon GPUs
Apple Silicon, NVIDIA through CUDA and AMD through ROCm are documented; Intel GPUs are not mentioned on the pages we captured, and Intel Macs are not supported.
4 quotes - Local API for other apps ×2 5
5/5 means: As 4, plus a second API (its own REST API or an Anthropic-compatible one) and official client libraries
OpenAI-compatible chat and embeddings endpoints served locally and on the network, Anthropic-compatible endpoints, and official Python and TypeScript SDKs.
4 quotes - Release recency ×2 5
5/5 means: Latest release less than 1 month old
The newest changelog entry, Bionic 1.2.0, is dated 5 November 2026.
1 quote
Evidence (24 quotes)
- 1
“Run llama.cpp (GGUF) or MLX models”
lmstudio.ai/docs/app · checked - 2
“Search & download functionality (via Hugging Face”
lmstudio.ai/docs/app · checked - 3
“You can also 'sideload' models (use models that were procured outside the app).”
lmstudio.ai/docs/app/offline · checked - 4
“Chip: Apple Silicon (M1/M2/M3/M4).”
lmstudio.ai/docs/app/system-requirements · checked - 5
“Intel-based Macs are currently not supported.”
lmstudio.ai/docs/app/system-requirements · checked - 6
“GPU: at least 4GB of dedicated VRAM is recommended.”
lmstudio.ai/docs/app/system-requirements · checked - 7
“CUDA 13 support on Windows ARM.”
lmstudio.ai/changelog · checked - 8
“Fixed ROCm runtime compatibility detection for AMD Radeon 8065S graphics.”
lmstudio.ai/changelog · checked - 9
“Send requests to Responses, Chat Completions (text and images), Completions, Embeddings, and Decisions endpoints.”
lmstudio.ai/docs/developer/openai-compat · checked - 10
“Serve local models on OpenAI-like endpoints, locally and on the network”
lmstudio.ai/docs/app · checked - 11
“Anthropic Compatibility Endpoints”
lmstudio.ai/docs/developer/openai-compat · checked - 12
“Use lmstudio-python to load models, generate text, embed content, and build agents.”
lmstudio.ai/docs/app · checked - 13
“Nothing you enter into LM Studio when chatting with LLMs leaves your device.”
lmstudio.ai/docs/app/offline · checked - 14
“When you search for models in the Discover tab, LM Studio makes network requests (e.g. to huggingface.co). Search will not work without internet connection.”
lmstudio.ai/docs/app/offline · checked - 15
“Can I use Bionic for free? Yes. The Free plan includes Bionic Agent and local model support.”
lmstudio.ai/pricing · checked - 16
“Bionic+ $ 20 USD / month Private, US-hosted OSS models.”
lmstudio.ai/pricing · checked - 17
“oMLX 0.1.0 engine for Apple Silicon Macs. Nov 5, 2026”
lmstudio.ai/changelog · checked - 18
“LM Studio is available for macOS, Windows, and Linux.”
lmstudio.ai/docs/app · checked - 19
“Pro $ 100 USD / month More included usage.”
lmstudio.ai/pricing · checked - 20
“Can I continue running models locally? Yes.”
lmstudio.ai/pricing · checked - 21
“In general, LM Studio does not require the internet in order to work. This includes core functions like chatting with models, chatting with documents, or running a local server, none of which require the internet.”
lmstudio.ai/docs/app/offline · checked - 22
“CPU: AVX2 instruction set support is required (for x64)”
lmstudio.ai/docs/app/system-requirements · checked - 23
“When you open LM Studio, the app updater will make a network request to check if there are any new updates available.”
lmstudio.ai/docs/app/offline · checked - 24
“5× usage limits”
lmstudio.ai/pricing · checked