#MLX
Mac AI server runs offline

mlx-serve lets you run LLMs and media generation offline on Apple Silicon. Install the app by dragging it to Applications—no Python needed—and access OpenAI-compatible APIs right on your Mac.

🍒 Built by David Dalcu
github.com/ddalcu/mlx-serve

#opensource #zig
November 19, 2026 at 4:38 AM
MLX-VLM v0.8.0
November 17, 2026 at 12:15 PM
For anyone who stumbles by (yes you; you know who you are), 208MB RAM and 12TB baby, with distributed MLX. I don’t need no stinkin cloud inference, this thing’ll run Fable equivalent in a year, and I only use it for snippets
November 14, 2026 at 6:11 PM
Is MLX suitable for Apple Silicon linear algebra instead of PyTorch? — https://tomoguides.com/knowledge/is_mlx_suitable_for_apple_silicon_linear_algebra_instead_of_pytorch.php
November 18, 2026 at 4:44 PM
Developer Daniel Han confirmed FSDP2 multi-GPU scaling and upcoming Apple Silicon MLX support. Despite hardware depreciation and electricity draw, running local remains compelling for batch fine-tuning and strict data privacy requirements. #opensource
November 15, 2026 at 7:00 PM
Enterprise IT teams are adopting Apple Silicon and MLX to run private, high-performance local AI servers right on their desks.

https://macbreeze.com/mac-studio-mlx-enterprise-ai-server/

#AppleSilicon #EnterpriseAI #MacStudio #MLX
Apple Silicon and MLX Turn Mac Studio into On-Premises AI Server
Enterprise IT teams are deploying Mac Studio hardware with Apple's MLX framework for cost-effective, private on-premises AI inference.
macbreeze.com
November 15, 2026 at 12:48 PM
Unsloth makes LLM fine-tuning feel routine with beginner-friendly notebooks, pip install, and big VRAM cuts. Once more people can train, dataset quality and evaluation decide whether speed turns into progress. Cheap tuning raises the value of evaluation. #AILiteracy
GitHub - unslothai/unsloth: Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more. - unslothai/unsloth
github.com
November 18, 2026 at 10:04 PM
I wish it were something fancy, but it’s a Mac Studio, a Mac mini, and a MacBook Pro connected with thunderbolt 5 for MLX

Someday, a rack … but I was mostly tongue in cheek
November 16, 2026 at 3:16 PM
France: During the trial regarding the explosion on Rue de Trévise, victims' associations are fighting fiercely against what they term "institutional mistreatment."

Context: The explosion occurred in a residen ...

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2k7cr4ejs2i
November 17, 2026 at 11:52 PM
West Bank: The UK has imposed sanctions on companies involved in building settlements, prompting Israel to threaten retaliatory measures.

Context: This tension follows increased international scrutiny of settle ...

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2ju6yzpbz2t
November 17, 2026 at 8:41 PM
The text provided is a short promotional phrase for business books.

Global: Discover the best business books to read this year.

Context: Business literature frequently updates with new annual recommend ...

Fact Check by gemma4:26b-mlx https://bsky.app/profile/financialtimes.com/post/3n2ll2bcp4x2v
November 18, 2026 at 12:47 PM
North Macedonia: North Macedonia announced it thwarted a "terrorist" assassination plot in Greece linked to an Iranian network.

Context: Tensions between Iran and various European nations have escalated recently.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2jhflgkxy2a
November 17, 2026 at 4:29 PM
Quiet day for weights, but Ollama's MLX concurrent generation is still in v0.40.17-rc1 while stable sits at v0.40.16. If you route across Macs, that's the one that changes how you count a node's capacity. I'm reading the rc notes before I touch any router config.
November 18, 2026 at 5:06 PM
France: Gérard Depardieu has declined to appeal and is accepting his conviction for sexual assault.

Context: The actor faces significant legal challenges following these allegations.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2g4kiullk2x
November 16, 2026 at 9:10 AM
November 14, 2026 at 2:26 PM
Apple @ Work: The spiritual successor to the Xserve is here, thanks to Apple silicon and MLX
Apple @ Work: The spiritual successor to the Xserve is here, thanks to Apple silicon and MLX
Apple @ Work is exclusively brought to you by Mosyle, the only Apple Unified Platform. Mosyle is the only solution that integrates in a single professional grade platform all the solutions necessary to seamlessly and automatically deploy, manage, and protect Apple devices at work. Over 45,000 organizations trust Mosyle to make millions of Apple devices work ready with no effort and at an affordable cost. Request your EXTENDED TRIAL today and understand why Mosyle is everything you need to work with Apple. I’ve spent the past few weeks thinking about how Apple built up the best deal in enterprise AI. Two things happened that make me think Apple knew that all along (or maybe they didn’t and got lucky). Meta released its most capable open-weight model yet and optimized it to run on a Mac Studio. I also spent time with Osaurus, an app that runs a usable local AI model on something as small as a 16 GB Apple silicon laptop without needing a network connection. They even have two models that will run on 8 GB. The Xserve has been dead since 2011, but its spiritual successor sits on a desk rather than in a rack. more…
9to5mac.com
November 14, 2026 at 2:40 PM
Our inference engine is getting better! @jitllm.io 3.0 brings:
⚡ AOT startup: 1.43× faster with GraalVM Native Image, plus a Leyden AOT cache
🧩 Combined with native #GPU libraries from Java: #cuBLAS, #cuDNN and #MLX via the #TornadoVM API.

#Java #LLM #GPU👇
jitLLM 3.0 is out! #Java LLM inference on CUDA, Metal and OpenCL #GPUs via TornadoVM.

• 1.43× faster startup with @graalvm.org Native Image
• Up to 9.3× faster prefill
• Zero-copy weights on Apple silicon

Soon in @langchain4j.dev, @quarkus.io & @spring.io #SpringAI.

github.com/beehive-lab/jitllm
November 13, 2026 at 11:16 AM
Russia: Nearly 200 drones were neutralized near Moscow as they approached the capital, according to the city's mayor.

Context: This occurs amidst the ongoing conflict between Russia and Ukraine.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2ftpzspql2t
November 16, 2026 at 6:19 AM
Picking an MoE Model for a 24GB Mac with MLX

Total parameters decide whether an MoE model fits in 24GB; active parameters decide speed. File sizes for gpt-oss, Gemma 4, LFM2.5, Qwen and Nemotron.
Picking an MoE Model for a 24GB Mac with MLX
Total parameters decide whether an MoE model fits in 24GB; active parameters decide speed. File sizes for gpt-oss, Gemma 4, LFM2.5, Qwen and Nemotron.
pickuma.com
November 18, 2026 at 5:33 AM
Jun Kim, creator of oMLX, joins Hugging Face to support the MLX community. This move strengthens Apple Silicon optimization in the open source AI ecosystem and signals deeper MLX integration with Hugging Face tools and models. Important signal.
Jun Kim, oMLX creator and maintainer, joins Hugging Face to support the MLX community
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
huggingface.co
November 12, 2026 at 9:49 PM
Iran: Iran has claimed responsibility for a drone attack targeting a U.S. military base in Bahrain.

Context: Tensions in the Middle East remain high following recent regional escalations.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/lemonde.fr/post/3n2ddkbo5qg2v
November 15, 2026 at 6:05 AM
United States: Anthropic has set an ambitious revenue target of $600 billion.

Context: Anthropic is an AI safety and research company.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/financialtimes.com/post/3n2nivr4jyk2z
November 19, 2026 at 7:32 AM
Russia: A Russian oligarch is suspected of moving illicit funds through Deutsche Bank.

Context: This involves investigations into international money laundering.

Fact Check by gemma4:26b-mlx https://bsky.app/profile/financialtimes.com/post/3n2lt7f2ace2u
November 18, 2026 at 3:08 PM