So far, three groundbreaking models in a week and another tomorrow. This model alone will help 95% of the world’s population talk to each other (for free). Probably one of the greatest gifts to humanity in a while!.

Another “wholly shit” day as Alibaba Qwen announce Qwen3-Coder. Just a day after they announced their Kimi-K2 quashing monster. This is another large model, the majority of us are no going to be running locally. However, you can, it’s totally open source. This new model, as the

Last week saw Kimi K2 not only take the open source crown but it crushed almost all of the OpenAI and Anthropic models. A week later the crown is taken back again by Alibaba’s Qwen3 with a new 235B model, ¼ the size of Kimi. Anyone can run these with total privacy, no one gets

I’m about to take my 18th train in Germany this year, 2nd today, 5th this week. Not one single train has been on time with delays ranging from 15 minutes to 2+ hours, averaging around 50 minutes. Deutsche Bahn is a joke, it makes British Rail look like Swiss trains.

I’m back in Munich, the organisers of @mlconference ask me last minute if I could teach a 2-day AI course. Stupidly I said yes and then spent all weekend writing exercises. Monday done, time for a beer 🍺

About to join Daniel’s rather excellent AI Happy Hour, somehow I manage to forget a drink though!! 🍺.

🤣🍺.

My colleague @AidfulAI found a prompt that seems to break the @ollama version of deepseek-r1 (8b-0528-qwen3) for Q4 and fp16. It works on MLX and base qwen3:b4. It’s…. “If a plant that doubles its size every day covers a lake in 30 days, how much time will it take for 2 plants.

For anyone wanting to update all their @ollama models in one line. ollama list | awk 'NR>1{print $1}' | xargs -n1 ollama pull.

Please join me on Tuesday evening (21h EU/ 20h UK time) for a chilled out discussion on everything AI, specifically open-source, local & private LLMs. This is run by @AidfulAI and we usually get @WolframRvnwlf on the space too.

Head up (literally), it looks like there’s severe geomagnetic storm coming our way (Earth) on the 2nd. This could result in some pretty spectacular aurora.

WTF is going on with Claude (Sonnet) 4???. It's being totally bloody "stupid" today, a good hour of fixing bugs and it's constantly adding new crap I didn't ask for, which then has bugs and we go full circle. @AnthropicAI Come on guys, what are you playing at?.

Some possibly useful performance stats on Qwen3. This following is on my M4 Max laptop running on battery with the prompt "Why is the sky blue?". Not exactly thorough but it might help a few people trying to decide which to download.

Wow, I’m stuck on a plane on my way to SFO and Alibaba have just released Qwen3! I can assure you I will be downloading it as soon as I get through immigration. Qwen2.5 remains one of the best, the stats on Qwen3 look amazing!.#MLX-LM first!

( ● ) Thinking. I was trying out OpenAI's codex just now, I'm used to claude and of course we have our own too. My second, relatively simple prompt simply bombed out with an error. Not a great start OpenAI, you have some catching up to do.

Not the greatest eclipse ever or the greatest photo but not bad for a phone. Watching the partial solar eclipse from Gare de Lyon in Paris.

It seems @serperapi is down, what’s going on guys? We’ve failed over to @serp_api.

I see @MistralAI's new Mistral-Small-3.1 is being posted on the @Huggingface MLX community. Just a week or so after Google posted Gemma3 we have a smaller, even more powerful model that's multi-modal with Vision too. While there are a few on @Ollama there's nothing official yet.

Ok, this will be a LOT of fun to play with. The Sesame demo of Maya was scarily impressive, I’ve registered and downloading. I can’t wait until @Prince_Canuma gets this MLX tuned!.