By Billy G.R. · 2 October 2026
Checked 2026-10-02. Author: Billy G.R. Retail prices move; the hardware catalog stores the Amazon snapshot, not a promise of stock.
Claim. LM Studio is a desktop app for running local models. The 2 October 2026 homepage leads with Bionic, its agent, and says the runtime uses MLX and llama.cpp. It is not an open-source project.
Method. lmstudio.ai. Treat “open source” claims about the app itself as wrong. The models you download have their own licenses.
As of.
Claim. Meta’s 10 August 2026 Muse Glimmer post named LM Studio as a partner path “in the coming days.” Confirm the model is actually listed in your build before you depend on it.
Method. research.meta.ai Muse Glimmer post. “Coming days” from August is not proof the button exists in October.
As of.
Claim. Leave 2–6 GB above the site-formula estimate for the app, the server, and a short KV cache. That range is editorial headroom, not an LM Studio spec.
Method. Formula example: 27B Q4 = 18.2 GB, so a 24 GB machine has about 6 GB before the OS takes its share. Do not treat 2–6 GB as measured on a reference board.
As of.
Methodology — parameter math, quantization bytes, and the source list.
LM Studio is a desktop application for downloading and running local models, with a local server other tools can call. On 2 October 2026 the homepage led with Bionic, which it calls LM Studio’s agent for open models: documents, code, and computer control, speech transcribed on device, runtime powered by MLX and llama.cpp. The same page mentions cloud services with a zero-data-retention claim and names large open models. I am not repeating those model names as hardware recommendations unless they are in our parameter list. The app is proprietary. Say “local desktop,” not “open source.” The weights you load are a separate license, often Apache-2.0 or something narrower. Read the card.
The useful trick, even if you never open Bionic, is the local HTTP server. Open WebUI, Continue, Aider, and OpenHands can all target an OpenAI-compatible base URL. You pick the GGUF in LM Studio. They send chat requests. That split keeps the agent UI replaceable when a project stalls. The LM Studio versus Ollama page is the choice between a desktop browser of models and a service you run headless. For an always-on box with no monitor, I still prefer Ollama. For a person who wants to see the model list, LM Studio.
Meta’s Muse Glimmer post, 10 August 2026, listed LM Studio among the apps that would be able to run Glimmer “in the coming days,” next to Ollama and Unsloth. That sentence is a plan from August. Open the app and see if the model is there before you write a guide for someone else that assumes a one-click download. Glimmer’s own memory table is on the Glimmer hardware page.
The site formula, params × bytes × 1.2 + 2, already includes a short-context allowance and 2 GB. LM Studio is a GUI plus a server plus, if you turn Bionic on, an agent loop. I leave another 2–6 GB in the plan so the app and a modest KV cache are not pretending to be free. That 2–6 GB is my editorial pad. It is not printed on lmstudio.ai as a requirement, and it is not an Apple specification. If the formula says 18.2 GB for Qwen3.8-27B at Q4, a 24 GB machine is the first one I trust, not a 20 GB fantasy card.
| Model at Q4 | Formula | Machine after the pad |
|---|---|---|
| K2-Horizon-7B | 6.2 GB | 8–12 GB. The pad fits. |
| Qwen3.8-27B | 18.2 GB | 24 GB, and do not also load a second model. |
| K2-32B or Muse Glimmer Q4 | 21.2 GB or 20 GB | 24 GB is the weights. 36 GB is the weights plus Bionic plus cache. |
| Same 30B at Q8 | 38 GB | 64 GB unified. A 36 GB Studio does not have the pad left. Calculator. |
Tool calling only works if the model and the runtime both support the schema you send. A 7B that “has tools” in a model card can still emit broken JSON. Step up in size before you blame the app. I am not listing a tok/s figure for Bionic.
LM Studio wants a machine you sit at. The Mac mini M5 Pro 24 GB (B0HGGHNQY6, $1,669.99 on 1 October 2026) is enough for a 7B agent and a tight 27B. The Studio M5 Max 36 GB (B0HGKSQMX6, $2,449) is the one that matches the headroom paragraph for a 32B. A 24 GB NVIDIA card is the Windows version of the same idea. The RTX 4090 catalog listing is B0BG94PS2F.
Mac Studio M5 Max 36 GB on Amazon
ASIN B0HGKSQMX6. Amazon Associates link. The live price is on that page, not locked in here.
ASIN B0BG94PS2F. Amazon Associates link. The live price is on that page, not locked in here.
A correction or a hardware question goes to the contact page.