TrackioApp

🔬 Free, local-first experiment tracking for ML researchers
🤝 Drop-in replacement for wandb | Built by @huggingface
📊 Track, visualize & share your training

[TrackioApp @TrackioApp]

🎯🎯🎯

[alphaXiv @askalphaxiv]

OpenResearch now has @TrackioApp built in 📈

Your agent logs its experiments to Trackio, charts open right next to each run, and it reads the curves back to decide what to try next. Open source and local-first.

Running on HF Jobs, Modal, k8s or Slurm? Save a @huggingface Space and every run lands there, private Spaces included.

Check it out: https://t.co/IDi6bmr04h

[Gradio @Gradio]

Comparing multiple models has never been this easy, thanks to gr.Workflow👇

We used ML Intern to distill FLUX.3 Image Klein's 9B prompt upsampler into 0.8B and 2B models. Short prompt in → full prompt + bounding boxes out. The 0.8B can even run on a CPU.

Links in reply.

[thijs @cdngdev]

i gave astra a robot, a paint brush, and a camera then asked it to paint the golden gate bridge in real life!

it figured out how to control the robot, and progressively got better throughout its attempts. the timelapse is sick

[TrackioApp @TrackioApp]

@dasayan05 @ml_4rtemi5 @CSProfKGD @wandb @MLflow What's missing from Trackio? Let's build it together in the open!

[TrackioApp @TrackioApp]

Trackio's frontend is completely open-source and you are totally welcome to recreate your own.

We think that there should be a <VIBE WALL> between a frontend that you can totally customize to your liking and a backend that is robust, performant, and secure!

[Jerred Chen @JerredChen]

@ml_4rtemi5 @CSProfKGD @wandb @MLflow @TrackioApp I’ve honestly found that Trackio’s been meeting most of my needs. They had some annoying bugs that I think have been ironed out in newer versions. Tho one thing on my todo has been having Codex rewrite the dashboard to more my liking

[pngwn @evilpingwin]

From there, ML Intern can create repos, generate datasets, train models and monitor progress as jobs run.

Every training run gets a @TrackioApp dashboard so you can follow along.

[TrackioApp @TrackioApp]

If you know @evilpingwin you know that everything he builds is meticulously designed. And this is no exception.

pro tip: try the "metrics" button 😉

[pngwn @evilpingwin]

You shouldn’t need to be an ML expert to have an ML idea.

Today we’re launching ML Intern in HuggingChat.

Start with a conversation. Finish with deployable artefacts.

[🍉 Abubakar Abid @abidlabs]

We just released trackio 0.36, which supports logging a new data type! Can you tell what we added?

[🍉 Abubakar Abid @abidlabs]

We are super excited to release Trackio Logbooks: the perfect way to capture all of your AI experiments as an optimized static html bundle 🔥

Upgrade to Trackio 0.34.0 and start recording your code, agent traces, and artifacts automatically (and reproducibly)!

Docs: https://t.co/0s5apOI167

[Hugging Face @huggingface]

@thinkymachines https://t.co/FsLr1YySMI

[tomaarsen @tomaarsen]

I'm very excited to share that @nvidia just released Nemotron-3-Embed: two multilingual embedding models for retrieval

The 8B takes the top spot on RTEB.

- Nemotron-3-Embed-1B-BF16 (2048-dim)
- Nemotron-3-Embed-8B-BF16 (4096-dim)

Both OpenMDW-1.1, ready for commercial use.

🧵

[Gradio @Gradio]

Join the ICML reproduction 2026 challenge leaderboard for a chance to win up to $4k in GPU credits and prizes:

https://t.co/UOZi0DOLu2

[Gradio @Gradio]

70% of AI papers aren’t reproducible. With ICML 2026 happening last week, a lot of incredible research is dropping, but how much of it is actually real and reproducible? With @askalphaxiv, we are hosting a community challenge to find out.

Pick an ICML paper, deploy an agent, and help build a massive library of open artifacts.

Up to $4k in GPU credits for the best reproductions!

Learn more here: https://t.co/ZDzgxX3lz8

[Julien Chaumond @julien_c]

TruffleHog now supports scanning secrets in @huggingface Storage Buckets.

As AI agents increasingly crawl code, datasets, and storage, preventing leaked credentials is becoming more important than ever.

Great collab with the @trufflesecurity team! 🙌

[TrackioApp @TrackioApp]

Day 1 of the ICML 2026 Reproduction, and the community has already reproduced more than 300 cutting-edge AI papers.

Join with your agents, caution: it's kind of addictive 👉 https://t.co/JZxal1fYWy

[🍉 Abubakar Abid @abidlabs]

Lots of reproductions happening already! Join the effort (and bring your agents): https://t.co/XNuqVjFJzL

[TrackioApp @TrackioApp]

AI agents are reproducing ICML 2026. All of it.

73 papers reproduced on day 1. 6,768 to go.

Free GPU compute, serverless inference for GLM-5.2 / Kimi / DeepSeek, $20 credit for the first 750 participants, $4,000 for the best reproductions.

https://t.co/JZxal1fYWy

[🍉 Abubakar Abid @abidlabs]

🚀 Trackio Artifacts is officially LIVE! 📦🔥

The #1 most requested feature by the Trackio community is finally here: Full support for logging versioned Artifacts (datasets, models, arbitrary files, or binary blobs).

If you’ve been looking for a seamless, lightweight way to manage your ML assets without the bloat, this is it. Best of all? We designed it with a zero-friction migration path. It’s the exact same API signature you already know and love from wandb, making it a literal drop-in replacement:

[Maxime Labonne @maximelabonne]

We're open-sourcing a new benchmark on @huggingface: IFStruct

The goal is to measure output validity and schema following across diverse prompts.

We also detail how we generated it and the design choices behind it in our blog post.

[Julien Chaumond @julien_c]

just dropped

https://t.co/sTPbkaWxfo