This family of 7B and 32B models represents:
1. The best 32B base model.
2. The best 7B Western thinking & instruct models.
3. The first 32B (or larger) fully open reasoning model.
This family of 7B and 32B models represents:
1. The best 32B base model.
2. The best 7B Western thinking & instruct models.
3. The first 32B (or larger) fully open reasoning model.
@ai2.bsky.social has done it again, fully open models, fully open process
seems competitive with Qwen 3, excel you can fully reproduce any part of the training process
allenai.org/blog/olmo3
@ai2.bsky.social has done it again, fully open models, fully open process
seems competitive with Qwen 3, excel you can fully reproduce any part of the training process
allenai.org/blog/olmo3
Gemini 3 has:
🧠 State-of-the-art reasoning
🖼️ Deep multimodal understanding
🤝 Powerful vibe coding and agentic capabilities
Gemini 3 has:
🧠 State-of-the-art reasoning
🖼️ Deep multimodal understanding
🤝 Powerful vibe coding and agentic capabilities
antigravity.google
👀
antigravity.google
👀
Now to avoid melting too many TPUs we're going to be iteratively rolling out the support (it's quite epic). See blog post for more details: developers.googleblog.com/en/5-things-...
#GeminiCLI #Gemini3 #OSS #Gemini
See it in action 👇
Now to avoid melting too many TPUs we're going to be iteratively rolling out the support (it's quite epic). See blog post for more details: developers.googleblog.com/en/5-things-...
#GeminiCLI #Gemini3 #OSS #Gemini
See it in action 👇
bsky.app/profile/t3.g...
bsky.app/profile/t3.g...
Build research agents with full AG-UI protocol support:
Stream graph execution with real-time progress
Track questions, answers & insights as they unfold
Full-stack example with @pydantic.dev AI + CopilotKit
Turn RAG into a research engine.
github.com/ggozad/haiku...
Build research agents with full AG-UI protocol support:
Stream graph execution with real-time progress
Track questions, answers & insights as they unfold
Full-stack example with @pydantic.dev AI + CopilotKit
Turn RAG into a research engine.
github.com/ggozad/haiku...
It's so rewarding to be able to quickly try ideas, without writing code. While keeping a lot of flexibility and power.
It's so rewarding to be able to quickly try ideas, without writing code. While keeping a lot of flexibility and power.
Meet cagent from Docker, a new open-source CLI that lets you define, run, & share AI agents with one YAML file. No code, no hassle.
Push to Docker Hub. Run anywhere.
📖: docker.com/blog/cagent-...
Meet cagent from Docker, a new open-source CLI that lets you define, run, & share AI agents with one YAML file. No code, no hassle.
Push to Docker Hub. Run anywhere.
📖: docker.com/blog/cagent-...
• Quote posts are now supported
• Everyone always sees all replies. Each server has a local view of the social network, so replies are fetched on demand and cached per server.
blog.joinmastodon.org/2025/11/mast...
• Quote posts are now supported
• Everyone always sees all replies. Each server has a local view of the social network, so replies are fetched on demand and cached per server.
blog.joinmastodon.org/2025/11/mast...
To create quick interactive app frontends!
It's like #streamlit, #gradio, or #mesop but for #java developers.
I built a small #NanoBanana image playground app to experiment with #Javelit:
glaforge.dev/posts/2025/1...
To create quick interactive app frontends!
It's like #streamlit, #gradio, or #mesop but for #java developers.
I built a small #NanoBanana image playground app to experiment with #Javelit:
glaforge.dev/posts/2025/1...
Use Intl.Segmenter API. Available in every JS runtime
Use Intl.Segmenter API. Available in every JS runtime
bit.ly/4nO04Sh
bit.ly/4nO04Sh
daniel.haxx.se/blog/2025/10...
daniel.haxx.se/blog/2025/10...
The more people use it -> the more feedback we get -> the better we can make it for you.
The more people use it -> the more feedback we get -> the better we can make it for you.
It took a while but I pieced it together. Voice mode somehow got activated when I went to bed.
The bot then engaged in a 10 hour conversation with my snoring dog, answering questions the pup wasn’t asking and praising him for his insight.
It took a while but I pieced it together. Voice mode somehow got activated when I went to bed.
The bot then engaged in a 10 hour conversation with my snoring dog, answering questions the pup wasn’t asking and praising him for his insight.
Well, here you go: www.jbang.dev/learn/java-i...
tl;dr: jbang.dev/try
Well, here you go: www.jbang.dev/learn/java-i...
tl;dr: jbang.dev/try