LM Studio
Desktop app for downloading and running local LLMs, with a built-in OpenAI-compatible server.
Liked as a low-friction desktop app for downloading and running local models123, with a built-in local server that plugs into coding agents like OpenCode456, and fast MLX support on Apple Silicon Macs.78 The main complaints are that it is closed source, with slow bug fixes and unclear telemetry91011, plus hardware quirks such as broken ROCm support on AMD integrated GPUs and out-of-memory crashes.1213
Why people recommend it
Local API server · 3
Its bundled localhost server, with OpenAI- and Anthropic-compatible endpoints, connects easily to agents and editors.
“I use LM Studio with the local server it ships and connect it to opencode. Takes 2 min to setup”
“LM Studio has a server that serves models over HTTP (localhost).”
Show 1 more
“I use LMStudio which supports OpenAI and Anthropic compatible api endpoints”
Easy setup · 3
Commenters call it the lowest-friction way to download, manage and chat with models in one app.
“I strongly recommend trying LM Studio - it's the lowest friction way to try out models”
“"LM Studio"[2] is not bad to start with as you'll be able to download, start/stop/manage and chat with your LLM model all in one place”
Show 1 more
“LM Studio (nicer app on the Mac)”
Mac MLX speed · 3
MLX support makes it faster on M-series Macs, and MLX builds of new models appear quickly.
“I use LM Studio with MLX support, which is specifically faster on M series Macs.”
“the excellent LM Studio”
Show 1 more
“I use LMStudio pretty regularly”
Daily driver · 3
Several commenters use it regularly for local models in coding and general questions.
“with Qwen3.8-27B in back in LM Studio, to be amazing and i use it daily”
“Recently I started playing with LM Studio and local models”
Show 1 more
“I've been using LM Studio for a while, this is a nice update.”
Faster than llama.cpp · 1
One commenter got about 70 tok/s after switching from a self-run llama.cpp setup that managed about 10 tok/s.
“I eventually switched to LM studio and the same model runs much better, like 70tk/s.”
Structured outputs · 1
It supports structured JSON-schema outputs, as llama.cpp does.
“llama.cpp supports structured outputs, as do some more consumer-oriented tools that wrap like LM Studio.”
Caveats and dissent
Closed source · 3
It is closed source, reported bugs often go unfixed, and its telemetry is unclear.
“Case to point is LM Studio. It's closed source, has bugs (duh!), and there's at least a GitHub issue tracker to report the bugs -- but then by and large nothing happens”
“I eventually switched to LM studio and the same model runs much better, like 70tk/s.”
Show 1 more
“open source alternative to the more widely known LM Studio”
Hardware quirks · 3
Users report that the ROCm runtime does not use the AMD GPU on Framework machines, along with out-of-memory crashes and slowdowns when a model is split across GPUs.
“Their ROCm runtime does not enable the the AMD GPU inside the Framework”
“I'm documenting my experience using aoe + OpenCode + LM Studio + GLM-4.7 Flash + Mac Mini M4Pro 64GB Mem”
Show 1 more
“offload some into system ram when using LM studio”
Missing features · 2
Users note that image-related parameters are not implemented and that there is no Android companion app.
“I got tired of LM Studio not implementing the image-related parameters”
“I love LM studio and this is super cool”
Token speed · 1
One user is considering switching away from it to get better token speed.
“I may have to switch away from LM-studio if I can improve token speed.”
Often compared with
1 moved from llama.cpp to LM Studio
1 moved from LM Studio to llama.cpp
1 moved from Ollama to LM Studio
Who recommends it in their own words
Threads where it came up
- LM Studio Bionic: the AI agent for open models 4 mentions
- Nativ: Run frontier open models locally on your Mac 3 mentions
- Qwen 3.6 27B is the sweet spot for local development 3 mentions
- Run Qwen3.8 27B locally: real numbers from my Mac Studio 2 mentions
- Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding? 2 mentions
- LM Studio 0.4.0 2 mentions
Data: lm-studio.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 43
“I used both lm studio and whatever the name of their new coding assistant”
“locally running models (at most 27B or 35B A3B or similar) on LM Studio”
“I use LM Studio with MLX support, which is specifically faster on M series Macs.”
“offload some into system ram when using LM studio”
“Using: LM Studio (primary), Lemonade”
“I've also heard okay things about LM Studio”
“with Qwen3.8-27B in back in LM Studio, to be amazing and i use it daily”
“I've heard good thing about LM Studio”
“I guess you can get it a lot more click to go with something like LM Studio though but I want to use it on a server”
“Others use LM Studio, oMLX, etc.”
“I got tired of LM Studio not implementing the image-related parameters”
“Their ROCm runtime does not enable the the AMD GPU inside the Framework”
“I used LM Studio and this model file”
“I eventually switched to LM studio and the same model runs much better, like 70tk/s.”
“I had a lot of chats in OpenRouter, google AI Studio and lm studio”
“Bionic provides a nice experience and makes trying new models out trivial.”
“"LM Studio"[2] is not bad to start with as you'll be able to download, start/stop/manage and chat with your LLM model all in one place”
“Case to point is LM Studio. It's closed source, has bugs (duh!), and there's at least a GitHub issue tracker to report the bugs -- but then by and large nothing happens”
“Recently I started playing with LM Studio and local models”
“I can't really tell why would I use this over like lm studio, jan.ai and such?”
“LM Studio (nicer app on the Mac)”
“the excellent LM Studio”
“I was already using LMStudio”
“I love LM studio and this is super cool”
“I love lm-studio, so cant wait to see how this goes. For local I normally use opencode + lmstudio.”
“I really love LM Studio”
“which is part of the reason I just switched to LM Studio from Ollama.”
“I use LM studio (local server), but can switch to a different tool.”
“I use LMStudio pretty regularly”
“I usually use LM Studio for this type of thing”
“I may have to switch away from LM-studio if I can improve token speed.”
“The main thing in LM studio (or whatever software you use”
“I can just hit my own APIs (via LM Studio) on the MBP over Tailscale.”
“I use LMStudio which supports OpenAI and Anthropic compatible api endpoints”
“I think he just uses LMStudio most of the time on his”
“llama.cpp supports structured outputs, as do some more consumer-oriented tools that wrap like LM Studio.”
“I strongly recommend trying LM Studio - it's the lowest friction way to try out models”
“with open code as the harness and llmster(LM Studio)”
“I use LM Studio to run and am using it via VSCode.”
“I use LM Studio with the local server it ships and connect it to opencode. Takes 2 min to setup”
“open source alternative to the more widely known LM Studio”
“LM Studio has a server that serves models over HTTP (localhost).”
“I'm documenting my experience using aoe + OpenCode + LM Studio + GLM-4.7 Flash + Mac Mini M4Pro 64GB Mem”