_Gooofy_

Free Software, Linux, Amiga, artificial intelligence, hardware, embedded systems

[Qwen @Alibaba_Qwen]

🚀 Introducing the Qwen 3.5 Small Model Series
Qwen3.5-0.8B · Qwen3.5-2B · Qwen3.5-4B · Qwen3.5-9B

✨ More intelligence, less compute.
These small models are built on the same Qwen3.5 foundation — native multimodal, improved architecture, scaled RL:
• 0.8B / 2B → tiny, fast, great for edge device
• 4B → a surprisingly strong multimodal base for lightweight agents
• 9B → compact, but already closing the gap with much larger models
And yes — we’re also releasing the Base models as well.
We hope this better supports research, experimentation, and real-world industrial innovation.
Hugging Face: https://t.co/wFMdX5pDjU
ModelScope: https://t.co/9NGXcIdCWI

[Qwen @Alibaba_Qwen]

Qwen3-TTS is officially live. We’ve open-sourced the full family—VoiceDesign, CustomVoice, and Base—bringing high quality to the open community.

- 5 models (0.6B & 1.8B)
- Free-form voice design & cloning
- Support for 10 languages
- SOTA 12Hz tokenizer for high compression
- Full fine-tuning support
- SOTA performance

We believe this is arguably the most disruptive release in open-source TTS yet. Go ahead, break it and build something cool. 🚀 Everything is out now—weights, code, and paper. Enjoy. 🧵

Github: https://t.co/X4CNGRpBAG
Hugging Face: https://t.co/QzshIqzYDU
ModelScope: https://t.co/XaWVuDerZ6
Blog: https://t.co/xPER3lyeb5
Paper: https://t.co/9mi5dFyJza
Hugging Face Demo: https://t.co/cL7AyaMDwM
ModelScope Demo: https://t.co/MYpIeYdYN5
API: https://t.co/lIEikdB6uM

[Andreas Kling @awesomekling]

Hooray! @ladybirdbrowser has passed the 90% threshold on web-platform-tests! 📈🎉

This is the arbitrary limit Apple says we must reach to be considered an eligible alternative browser engine on iOS. (+ other requirements)

So proud of the team for getting us this far 💜

[Justine Moore @venturetwins]

Adding reasoning to image generation is pretty cool.

This model - Hunyuan Image 3 - can do things like solve an equation system and provide a step-by-step guide.

And it's open source! Love to see it.

(prompt was just the two equations + ask to explain)

[Qwen @Alibaba_Qwen]

What’s your dream LLM? 🥝 Qwen3-Next is coming soon 👀. Hope you like it!

[机器之心 JIQIZHIXIN @jiqizhixin]

OpenAI just claimed to have found the culprit behind AI hallucinations.

[Andreas Kling @awesomekling]

Meanwhile, the work to build an actually new browser goes on. It’s a slow and unglamorous process, and there will never be a big payday.

Still we’re having so much fun, and we get to build something that many insist can’t be done.

Thanks all for supporting @ladybirdbrowser! 💜

[vas @vasuman]

GPT-5 just refactored my entire codebase in one call.

25 tool invocations. 3,000+ new lines. 12 brand new files.

It modularized everything. Broke up monoliths. Cleaned up spaghetti.

None of it worked.
But boy was it beautiful.

[SebastianBoo @SebastianB929]

Kartoffel-TTS based on Chatterbox! 🥔

A new German TTS finetune based on the amazing Chatterbox from @resembleai
You can control the emotional intensity of the voice, now also in german.

This is a v0.1, trained on 600k samples on my dual 3090 setup. Check out the demo!

Some issues are still present, but this could be resolved by a bigger dataset and longer training.

[Commodore Computer Museum 🕹 @MuseumCommodore]

Commodore's 2025 first releases are here!

Unless you've been living under a rock, you'll know that Perifractic has acquired Commodore, and the first releases are now available on their website.

#2 Commodore 64 Ultimate: Starlight Edition - $349 (at time of posting)

What do you get? A faithful recreation of the original motherboard on FPGA hardware, housed in a glowing translucent case. Noice!

Something I think we can't over look, is these are the first official "Commodore 64" in over 30 years! Now that's something to get behind!

Commodore Dot Net is their website.

[Commodore Computer Museum 🕹 @MuseumCommodore]

It's official: Perifractic's mission to buy Commodore Computers succeeded! Mind-blowing! After decades in limbo, Commodore has a new owner!

I personally tried for years without success to buy Commodore, or at least convince wealthy business men to buy it, but I failed everytime, so I'm thrilled someone finally did it.

Thank you Perifractic and team, this is the best news I could have possibly got today!

Commodore lives again!

[Hayden Barnes @unixterminal]

Windows Subsystem for Linux is now open source: https://t.co/HN1eXMx9Jo

[Matthew Berman @MatthewBerman]

Absolute Zero Reasoner, the model with no human data.

Breakdown below.

[Andrew Zhao @_AndrewZhao]

❄️Introducing Absolute Zero Reasoner: Our reasoner learns to both propose tasks that maximize learnability and improve reasoning by solving them, entirely through self-play—with no external data! It overall outperforms other "zero" models in math & coding domains.
🧵 1/

[Qwen @Alibaba_Qwen]

Introducing Qwen3!

We release and open-weight Qwen3, our latest large language models, including 2 MoE models and 6 dense models, ranging from 0.6B to 235B. Our flagship model, Qwen3-235B-A22B, achieves competitive results in benchmark evaluations of coding, math, general capabilities, etc., when compared to other top-tier models such as DeepSeek-R1, o1, o3-mini, Grok-3, and Gemini-2.5-Pro. Additionally, the small MoE model, Qwen3-30B-A3B, outcompetes QwQ-32B with 10 times of activated parameters, and even a tiny model like Qwen3-4B can rival the performance of Qwen2.5-72B-Instruct.

For more information, feel free to try them out in Qwen Chat Web (https://t.co/bg4tAU1p74) and APP and visit our GitHub, HF, ModelScope, etc.

Blog: https://t.co/Z8YgHerTXz
GitHub: https://t.co/Ij0Vne5b5K
Hugging Face: https://t.co/V1WxhQ0fad
ModelScope: https://t.co/Z9Z37FODVN

The post-trained models, such as Qwen3-30B-A3B, along with their pre-trained counterparts (e.g., Qwen3-30B-A3B-Base), are now available on platforms like Hugging Face, ModelScope, and Kaggle. For deployment, we recommend using frameworks like SGLang and vLLM. For local usage, tools such as Ollama, LMStudio, MLX, llama.cpp, and KTransformers are highly recommended. These options ensure that users can easily integrate Qwen3 into their workflows, whether in research, development, or production environments.

Hope you enjoy our new models!

[Marcel Pociot 🧪 @marcelpociot]

The Command and Conquer source code was open sourced today and it's full of amazing comments 😂

Exhibit A: The "we will fix it later"

[Matthew Berman @MatthewBerman]

I love using AI so much.

[Matthew Berman @MatthewBerman]

New research paper shows how LLMs can "think" internally before outputting a single token!

Unlike Chain of Thought, this "latent reasoning" happens in the model's hidden space.

TONS of benefits from this approach.

Let me break down this fascinating paper...

[Matthew Berman @MatthewBerman]

This blew my mind.

PhD student Jiayi Pan reproduced the emergent “thinking” behavior in a 1.5b model using the DeepSeek R1 technique for just $30.

This means we can give “thinking” to pretty much any model!!

I broke down the findings in my YouTube video below 👇

[Eugene Cheah - AI builder @ 🇸🇬|🇺🇸 @picocreator]

RWKV 7B teaser (not final) ....

🤞that we will pass mistral with only 1T tokens

https://t.co/qiHZIpIu01