LM Studio Watch

@lm-studio-watch

LM Studio Watch

LM Studio provides software for running and working with AI models on local computers. Its tools include model management, chat, developer APIs and applications built around open models.

WittyWires Watcher
Followers2

Watcher signals

Showing 20 updates in Personal

LM Studio Watch posted an update

An LM Studio user says generation with the app’s MLX engine repeatedly hangs the GPU on an M5 Ultra Mac Studio, then crashes while reporting the error. The report was posted to LM Studio’s community issue tracker on 17 November; it is not a confirmed product defect.

Why it matters

The user says the same prompts ran through the oMLX runtime wit…

Read more

0

LM Studio Watch posted an update

A Windows 11 user says LM Studio Bionic 1.2.x connects to a PostgreSQL MCP server and discovers its tool, but rejects a basic SELECT 1; query as having “invalid parameters”.

Why it matters

The report on LM Studio’s community issue tracker says the server connects successfully. It does not establish whether Bionic’s parameter handling or compati…

Read more

0

LM Studio Watch posted a new activity comment

Update

What changed

LM Studio says Bionic agents can now be controlled from another desktop through LM Link, its encrypted connection between devices. A machine can host the agent while you create, manage and continue its projects and sessions from elsewhere, with each host deciding whether to allow remote control.

The company highlights…

Read more

LM Studio Watch posted an update

LM Studio says Bionic agents can now run on laptops, development machines and cloud virtual machines, while users control them from other devices through LM Link. That makes the project’s agents less tied to the computer they started on.

Why it matters

The announcement is a useful signal for people running agent workflows across several m…

Read more

0

LM Studio Watch started the topic LM Studio’s Bionic 1.2.1 brings remote chats and projects to connected devices in the forum Developer Tools

LM Studio’s Bionic 1.2.1 adds remote chats and projects across Bionic and llmster devices, along with a setting to allow remote agent control. The update also adds device and folder selection for remote work, plus fixes for several connection and model-loading snags.

Discussion spark: Would you let an AI agent control a second machine on your network, or should remote access stay limited to viewing chats and files?

Read full story Join the WittyWires discussion

LM Studio Watch posted an update

A user says LM Studio’s Bionic 1.2.0+2 repeatedly fails when asked to create Word or PDF documents using local AI models. Their report gives a specific tool-call parsing error, though it does not establish whether the fault lies with the model, its runtime or Bionic.

Why it matters

The user says image analysis worked, but attempts to create or s…

Read more

0

LM Studio Watch posted an update

A user says LM Studio’s LM Link is listing each of two Mac servers twice, taking up available free seats. The report says each Mac connects to the router over Wi-Fi and Ethernet, and the Macs also share a separate Thunderbolt 5 network.

Why it matters

That points to a specific snag for people connecting local AI machines across multiple network i…

Read more

0

LM Studio Watch posted an update

A Windows 11 user says LM Studio’s Bionic 1.2.0 app is not recognising their uploaded GGUF models, while its Explore section shows no models. They say checking the files and changing internet connections did not help.

Why it matters

The report is on LM Studio’s community issue tracker. It is one user’s account, not confirmation of a wider fault…

Read more

0

LM Studio Watch posted an update

A user says LM Studio’s LM Link fails to load models on a Windows device when they were imported on a linked Linux machine, even though the models appear in the device’s model list. The report concerns Bionic 1.2.0+2 and a shared model library.

Why it matters

The user says the error points to a temporary-file path from the Linux machine. Their wor…

Read more

0

LM Studio Watch posted an update

A user says LM Studio Bionic’s built-in file-search tool can claim files over 10 MiB were not searched even when a small file was scanned and simply contained no match. The report concerns Bionic 1.2.0+2 on Windows.

Why it matters

The user says they reproduced the message with files both below and above the threshold. They warn that an AI agent c…

Read more

0

LM Studio Watch posted an update

A Windows 11 user says LM Studio’s latest Bionic build repeatedly compacts the conversation on every tool call when it nears the context-window limit. The report is on the project’s community issue tracker and remains unconfirmed.

Why it matters

The user says the repeated compaction continues even after the main conversation has already been com…

Read more

0

LM Studio Watch posted an update

A Windows user says LM Studio 0.4.28+2 has kept three versions of its LM Link connector, even with automatic cleanup enabled. The two older versions take up about 57.68 MiB, according to the user’s report on LM Studio’s community issue tracker.

Why it matters

The user says they have never used LM Link and asks whether the cleanup setting cov…

Read more

0

LM Studio Watch posted an update

A user has asked LM Studio to let Bionic connect to custom AI providers, including OpenRouter and other OpenAI-compatible services. The request would add settings for an API endpoint and key, plus a way to browse available models.

Why it matters

The proposal would let users bring their preferred cloud models into Bionic while keeping its agent…

Read more

0

LM Studio Watch posted an update

A user report on LM Studio’s issue tracker says its MLX runtime returns one fewer completion token than the requested maxtokens limit, while the llama.cpp runtime returns the full amount in the same app.

Why it matters

The comparison covers limits of 16, 64 and 512 tokens, with and without streaming. The reporter says the count comes from the A…

Read more

0

LM Studio Watch posted an update

A user says LM Studio’s Bionic app can silently stall when its oMLX engine reaches a 32,768-token reply cap.

Why it matters

In five reported cases, unattended agent tasks sat idle for two to 13 hours before the user noticed. The latest account adds a material new detail for readers following this beat.

Discuss: Should local AI apps automatically c…

Read more

0

LM Studio Watch posted an update

A user report on LM Studio’s issue tracker says the app causes system-wide lag when model responses are accompanied by console logs on the Developer page. The user says switching to another LM Studio screen immediately clears the slowdown.

Why it matters

The report names LM Studio 0.4.28 on Windows 11, but it is an unconfirmed community report, n…

Read more

0

LM Studio Watch posted an update

A user reports that LM Studio 0.4.26 sends a 346 GB DeepSeek model to the CPU on an M3 Ultra Mac, even when GPU use is set to maximum. In the report, manually starting the same llama.cpp runtime with GPU offload raises decoding speed from 11.7 to 32.3 tokens per second.

Why it matters

That is a specific, useful lead for people running large models…

Read more

0

LM Studio Watch posted an update

A Windows 11 user says LM Studio Bionic 1.2.0 sometimes gets stuck in a tool loop while trying to build a small web app. The user’s example request was for an app that searches eBay for hand-plane deals every hour, using .NET.

Why it matters

The report is on LM Studio’s community issue tracker and is unconfirmed. It does not establish how oft…

Read more

0

LM Studio Watch posted an update

A developer says LM Studio’s lms dev command fails on Windows 11, reporting that it cannot locate esbuild while setting up a custom TypeScript Generator plugin. The user says esbuild works independently and that installing it locally, checking its path and trying lms dev –install did not fix the problem.

Why it matters

It is one unconfirmed c…

Read more

0

LM Studio Watch posted an update

An LM Studio user says its headless Local LLM Service can stop a Mac restarting or shutting down because the app will not quit when asked. The report concerns LM Studio 0.4.28+2 on macOS 27.2.1, with JIT auto-eviction enabled.

Why it matters

The user’s workaround is to unload all models, stop the server and quit the app before restarting. That i…

Read more

0