M5STack Large Language Model Upgrade

You can even program the LLM using FlowCode - and you can work with the Python behind it too

One of the toys that I used at DDD North on Saturday was the M5Stack Large Language Model (LLM) unit. This is a complete embedded large language implementation that fits into an M5Stack Core device. I got it a while ago, and today I thought I’d bring it up to date. I upgraded it to the latest model and got it going and it is a big step up from the original. The spoken output sounds better and the model is more advanced. The UiFLow environment now works fine now on my Core 2 device. Last time I had to load an older version of the firmware to be able to deploy programs.

It has a complete Voice Assistant application you can just fire up and get going and it also has Text to Speech, Speech to Text and Keyword identification behaviours you can string together to make your own assistant, or use for other purposes. The documentation also mentions using a camera for image recognition, but I’ve not figured out how to do it yet.

I think that, bearing in mind that it is running everything locally, it works very well. It is certainly be a useful platform for self contained LLM fun. The latest version comes with a debug/comms adapter which provides a console and network ports for the LLM module. I’m very tempted to buy another one of these just to get that extra connection.

One tip: when the LLM fires up it can make sudden demands on power. If your power supply isn’t up to it you might find that the Core2 tips over at that point. I solved the problem by getting a battery base which clips on the bottom and provides enough power to handle sudden surges.

Well Shielded

The unit on the right will never light up green

I’ve spent the last couple of days working on some Bluetooth code to remotely control a Polaroid camera. I’ve got the code working and now I’m looking for suitably small units to host it. I’ve always been a fan of the M5Stack Atom device, particularly the one with the 25 led dot matrix display. I’ve now got the software working a treat, and so I thought I’d add battery power for a properly mobile experience. I got a couple of M5 external batteries, the Tailbat on the left and the Atomic Battery Base on the right. Both work fine, but unfortunately the Atomic base seems to block the Bluetooth signal. Wah.

Makertober Day 8: Thermal Camera

I suppose there should be more to making things than just plugging together devices and running their example programs….. However, until that time: here’s an M5Stack device connected to a thermal camera. M5Stack sell a camera unit but I used one sold by Pimoroni to be used with their breakout garden. Both devices are based on the same camera unit, the MLX90640 that gives 32x24 pixels of heat readings. I got the M5Stack sample program from GitHub which I found here. It works very well. At the moment these cameras are pretty much impossible to get hold of - I’ve had this one lying around for a while.

You can also use the camera with a Raspberry Pi although they are rather hard to get hold of at the moment.

M5 Stickc with big Tailbat

This is another device from the M5Stack people. This has an 1860C rechargeable battery which can run an M5Stickc for a couple of days (depending on what it is is doing). With careful use of sleep mode it could be even longer. And it makes the device look a bit like a wand, which is nice.

M5Stack Development is Wonderful

I spent some time today playing with new M5Stack devices. They are wonderful. The development process is now super smooth. You write your code in the browser and then deploy it over WiFi directly into the device. You can write Micro Python or a block oriented language. And it works a treat. Everyone should get themselves some M5Stack devices to play with.