Herbo

AI Engineer Living The Dream.

[Herbo @herb_otto]

@MacRumors Walkie Talkie is extremely useful and works great for my wife and I over LTE - at least make it a downloadable app. @Apple @tim_cook

[Gunther Eagleman™ @GuntherEagleman]

History has a feeling before it has a headline.

You can feel it in the air.

A nation rediscovering its strength.

A people rediscovering their confidence.

An America that refuses to settle for decline.

The next chapter is being written right now!

[Herbo @herb_otto]

@EneFlorian17 Prompt: write a bubble sort in python.
Speed: 63.8 tok/sec
I used oMLX to serve the model.

[Leading Report @LeadingReport]

Virginia Supreme Court overturns congressional map that would have left only 1 GOP congressional seat.

[Herbo @herb_otto]

Been using this MLX model server lately, tested on M4 and M5 Macs - quite nice. Does local caching, has direct downloading from huggingface and provides decent model level control. I no longer use LM Studio or Ollama, this covers all my bases.

https://t.co/wrtlZOBCOA

[Ashok Elluswamy @aelluswamy]

You are unequivocally safer using FSD!

[webAI @thewebAI]

Our first open-source release.

YOLO26-MLX, native YOLO26 on Apple Silicon. No PyTorch. No external GPU.

Up to 2.6x faster inference. Up to 1.7x faster training. Accuracy within 0.2% of official results.

It won't be the last.

[ollama @ollama]

Ollama is now updated to run the fastest on Apple silicon, powered by MLX, Apple's machine learning framework.

This change unlocks much faster performance to accelerate demanding work on macOS:

- Personal assistants like OpenClaw
- Coding agents like Claude Code, OpenCode, or Codex

[Herbo @herb_otto]

@ah20im At times I found Low to be superior to Xhigh. Seems to get “stuck in a rut” often when the local context grows

[Herbo @herb_otto]

Qwen3-Coder-Next-4bit running on 64GB M4 Pro bringing in 60 tokens per second. Not too shabby mlx-lm and LM Studio servers. Local vibe coding with opencode or aider is doable!

[Herbo @herb_otto]

Using the Qwen3-Next-80B-A3B-Instruct-4bit on a Mac Mini M4 Pro with 64GB unified memory. Nice model for local vibe coding. MLX Server.

[Herbo @herb_otto]

Chesapeake Bay started to freeze last night #Chesapeake #winterstorm2026

[Herbo @herb_otto]

@elonmusk I’m retired and another monthly bill for our 2 Tesla’s is not practical. Yet I’m one who benefits the most due to my aging. If fees increase I will be forced to unsubscribe in the future when I need it the most. Most likely look elsewhere.

[Herbo @herb_otto]

2026 Moon - #seestar50 #Astrophotography #fullmoon #NewYear2026

[Locally AI - Local AI Chat @LocallyAIApp]

Run the latest Ministral 3 models by @MistralAI on macOS.

Running privately on-device and optimized for Apple Silicon with MLX.

[The Wall Street Journal @WSJ]

The first time I let the vehicle take control, something shifted. As the car guided itself through traffic, I felt tension melt away https://t.co/cQUhZQOP5x

[Tesla @Tesla]

FSD has many times more driving experience than any human driver – our fleet is trained on >100 years of data & collectively experiences a lifetime of driving scenarios every 10 minutes

We validate across billions of miles of real-world driving, including various road types, lighting & weather conditions, traffic patterns, speeds & geographies

[Andrej Karpathy @karpathy]

I took delivery of a beautiful new shiny HW4 Tesla Model X today, so I immediately took it out for an FSD test drive, a bit like I used to do almost daily for 5 years. Basically... I'm amazed - it drives really, really well, smooth, confident, noticeably better than what I'm used to on HW3 (my previous car) and eons ahead of the version I remember driving up highway 280 on my first day at Tesla ~9 years ago, where I had to intervene every time the road mildly curved or sloped. (note this is v13, my car hasn't been offered the latest v14 yet)

On the highway, I felt like a passenger in some super high tech Maglev train pod - the car is locked in the center of the lane while I'm looking out from Model X's higher vantage point and its panoramic front window, listening to the (incredible) sound system, or chatting with Grok. On city streets, the car casually handled a number of tricky scenarios that I remember losing sleep over just a few years ago. It negotiated incoming cars in tight lanes, it gracefully went around construction and temporarily in-lane stationary cars, it correctly timed tricky left turns with incoming traffic from both sides, it gracefully gave way to the car that went out of order in the 4-way stop sign, it found a way to squeeze into a bumper to bumper traffic to make its turn, it overtook the bus that was loading passengers but still stopped for the stop sign that was blocked by the bus, and at the end of the route it circled around a parking lot, found a spot and... parked. Basically a flawless drive.

For context, I'm used to going out for a brief test drive around the neighborhood to return with 20 clips of things that could be improved. It's new for me to do just that and exactly like I used to, but come back with nothing. Perfect drive, no notes. I expect there's still more work for the team in the long march of 9s, but it's just so cool to see that we're beyond finding issues on any individual ~1 hour drive around the neighborhood, you actually have to go to the fleet and mine them. Back then, I processed the incredible promise of vehicle autonomy at scale (in the fully scaleable, vision only, end-to-end Tesla way) only intellectually, but now it is possible to feel it intuitively too if you just go out for a drive. Wait, of course surround video stream at 60Hz processed by a fully dedicated "driving brain" neural net will work, and it will be so much better and safer than a human driver. Did anyone else think otherwise?

I also watched @aelluswamy 's new ICCV25 talk last week (https://t.co/RdaM23kvez) that hints at some of the recent under the hood technical components driving this progress. Sensor streams (videos, maps, kinematics, audio, ...) over long contexts (e.g. ~30 seconds) go into a big neural net, steering/acceleration comes out, optionally with visualization auxiliary data. This is the dream of the complete Software 1.0 -> Software 2.0 re-write that scales fully with data streaming from millions of cars in the fleet and the compute capacity of your chip, not some engineer's clever new DoubleParkedCarHandler C++ abstraction with undefined test-time characteristics of memory and runtime. There's a lot more hints in the video on where things are going with the emerging "robotics+AI at scale stack". World reconstructors, world simulators "dreaming" dynamics, RL, all of these components general, foundational, neural net based, how the car is really just one kind of robot... are people getting this yet?

Huge congrats to the team - you're building magic objects of the future, you rock! And I love my car <3.