The bee in wandb is buzzing around at Fully Connected London.

That's a wrap on Day 1 of Fully Connected London. The atmosphere was incredible. Today was all about building. We saw developers fine-tuning LLMs, creating research agents, and architecting multi-model systems. Getting to connect with builders like this is why we do what we do.

I'm so excited to share that @CoreWeave is buying @marimo_io. The Marimo team has built a much loved tool that we've used and admired for a long time and we're all really excited to work together.

We have big news to share: marimo is joining @CoreWeave! We're doubling down on open-source and scaling molab with serious compute Our mission is the same: to build the world’s best open-source notebook for working with data Read the full announcement:

@tryhardfifi @weave_wb I did a project with weave and it’s really easy to use right, don’t know why there is no hype

We built ¡alacard! this weekend at WeaveHacks 2 hackathon to close out @Techweek_ with Weights & Biases @weights_biases and Google Cloud @GoogleCloudTech. It's a cross-provider cookbook hub for AI. Mix AI models and tools to generate runnable, traceable, self improving notebooks.

A glimpse into what the last 2 mins of a hackathon looks like right before submissions.

We've started a great tradition at CoreWeave of shipping an integrated new product weeks after acquisition - congrats @OpenPipeAI on the serverless RL launch!

RL X-mas came early. 🎄 For too long, building powerful AI agents with Reinforcement Learning has been blocked by GPU scarcity and complex infrastructure. That ends today. Introducing Serverless RL from wandb, powered by @CoreWeave! We're making RL accessible to all.

“People think you can throw 100,000 humans at a problem… A lot of competitors are incredibly manual—basically body shops… literally dumping data into spreadsheets… there’s almost no technology going on.” - @echen, CEO & Founder of @HelloSurgeAI Full episode below!

Had a great conversation with @l2k on Gradient Dissent about what's actually happening in post-training right now. We dug into why LMSYS is setting the industry back - researchers are literally being forced to make models hallucinate more to climb the leaderboard...

Evaluating LLMs in @weave_wb just got a whole lot easier! Now you can run your evals right in the UI, without writing a single line of code. Select datasets, pick models, include LLM-as-judge scorers, and hit Run. Simple as that.

Our W&B Inference service just got a major upgrade. @Zai_org's GLM-4.5 is now live on our platform! This ~300B param open model excels in reasoning, coding & agentic tasks via multi-token prediction & RL. See below to find out how you can get $50 in credits to try it out!

CoreWeave is buying @OpenPipeAI - I'm a big fan of the @corbtt and the ART RL library and excited to work together!

We shipped 3 quality-of-life updates: - Markdown rendering in Playground, more useful for looking at markdown outputs when comparing LLMs. - New Trace Graph View for visualizing trace trees. - Time-to-first-token metric to track LLM latency. All aimed at faster debugging.

BIG LAUNCH: @deepseek_ai's V3.1 is now live on W&B Inference! One model, two modes: toggle between high-speed 'Non-Think' & deep 'Think'. Priced at just $0.55/$1.65 per 1M tokens, it’s a game-changer for building intelligent agents. Want $50 in free credits? Details below.

Or if you're brave just run curl -sfL https://t.co/vaIifyCKWq | sh

You can find it at

My cofounder @vanpelt but a new open source product called catnip. It makes it easy to run @claude_code on your project in many parallel docker containers at the same time. It handles port forwarding really nicely. I've been using it for the past couple weeks and loving it.

"CEOs have assistants, chiefs of staff, and coaches. AI will give that to everyone—co-workers, assistants, and coaches powered by agents. It’s your own dream team, helping you do 90% of your work." – @jainarvind, CEO & Founder, @Glean