Thibault Lecailtel

Machine Learning @Ubisoft

[Thomas Wolf @Thom_Wolf]

Wow - this is the model currently sitting at #5 on Artificial Analysis - now open-source

Perf on AA somewhere between Claude 4.5 Sonnet and Grok4 fast. Topping models like Gemini 2.5 pro and Opus 4.1

[hardmaru @hardmaru]

Evolution Strategies can be applied at scale to fine-tune LLMs, and outperforms PPO and GRPO in many model settings!

Fantastic paper “Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning” by @yule_gan, Risto Miikkulainen and team.

https://t.co/CEyX6Z5ulG

[Thomas Wolf @Thom_Wolf]

Thrilled to finally share what we've been working on for months at @huggingface 🤝@pollenrobotics

Our first robot: Reachy Mini

A dream come true: cute and low priced, hackable yet easy to use, powered by open-source and the infinite community.

Tiny price, small size, huge possibilities. A robot built to code, learn, share with AI builders of all ages, all around the globe, using the latest vision, speech and text AI model. A first robot for today's and tomorrow's AI builders.

Read more and order now at https://t.co/UpjRipw5tP

First deliveries expected right after the summer.

[Sakana AI @SakanaAILabs]

Introducing The Darwin Gödel Machine: AI that improves itself by rewriting its own code

https://t.co/wEEB4LGPr0

The Darwin Gödel Machine (DGM) is a self-improving agent that can modify its own code. Inspired by evolution, we maintain an expanding lineage of agent variants, allowing for open-ended exploration of the vast design space of such “self-improving” agents.

Modern agentic systems, while powerful, remain static—once deployed, their intelligence remains fixed. We believe continuous self-improvement is key to the development of stronger AI capabilities. Our Darwin Gödel Machine is built from the ground up to enable AI systems that can learn and evolve their own capabilities over time, just as humans do.

On SWE-bench, DGM automatically improved its performance from 20.0% to 50.0%. Similarly, on Polyglot, the DGM increased its success rate from an initial 14.2% to 30.7%, significantly outperforming representative hand-designed agents.

Learn more about our approach in our technical report: https://t.co/kDNWFgCI6C

This work was done in collaboration with Jeff Clune (@jeffclune)’s lab at UBC, and led by his PhD students Jenny Zhang (@jennyzhangzt) and Shengran Hu (@shengranhu), together with Cong Lu (@cong_ml) and Robert Lange (@RobertTLange).

Code: https://t.co/RcYLd22TB5

[Mistral AI @MistralAI]

Meet Document AI, our end-to-end document processing solution powered by the world’s best OCR model!

https://t.co/wGjAGBMmZ0

[Mistral AI @MistralAI]

Meet Devstral, our SOTA open model designed specifically for coding agents and developed with @allhands_ai

https://t.co/LwDJ04zapf

[Sherjil Ozair @sherjilozair]

Today I'm launching my new company @GeneralAgentsCo and our first product.

Introducing Ace: The First Realtime Computer Autopilot

Ace is not a chatbot. Ace performs tasks for you.

On your computer. Using your mouse and keyboard.

At superhuman speeds!

[Jim Fan @DrJimFan]

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight:

- Real humanoid teleoperation data.
- Large-scale simulation data: we are open-sourcing 300K+ trajectories!
- Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”!
- Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos.

GR00T N1 is a single end-to-end neural net, from photons to actions:

- Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions.
- Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2.

We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings.

While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right.

Let’s solve robotics, together, one token at a time.

Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

[Mistral AI @MistralAI]

Introducing Mistral Small 3.1.
Multimodal, Apache 2.0, outperforms Gemma 3 and GPT 4o-mini.
https://t.co/BHLAAaKZ9w

[AI at Meta @AIatMeta]

Introducing Meta Video Seal: a state-of-the art comprehensive framework for neural video watermarking. Try the demo ➡️ https://t.co/oX7ZlbP5nc

[Zhou Xian @zhou_xian_]

Everything you love about generative models — now powered by real physics!

Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications.

Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: https://t.co/bEkIlCKqdf).

The Genesis physics engine and simulation platform is fully open source at https://t.co/DhBv7NdyqH. We'll gradually roll out access to our generative framework in the near future.

Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism.

We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications.

Open Source Code: https://t.co/DhBv7NdyqH
Project webpage: https://t.co/SBNyhFB0yn

Documentation: https://t.co/3yuBoaealV

1/n

[Ruud van der Linden @RuudNL]

Sora v2 release is impending:

* 1-minute video outputs
* text-to-video
* text+image-to-video
* text+video-to-video

OpenAI's Chad Nelson showed this at the C21Media Keynote in London. And he said we will see it very very soon, as @sama has foreshadowed.

[Google DeepMind @GoogleDeepMind]

Introducing Genie 2: our AI model that can create an endless variety of playable 3D worlds - all from a single image. 🖼️

These types of large-scale foundation world models could enable future agents to be trained and evaluated in an endless number of virtual environments. → https://t.co/1dzB2BUlWo

[Guillermo Rauch @rauchg]

AI vectorization works so freaking well
left: low quality jpg (input)
right: high quality svg (output)

[Genmo @genmoai]

Introducing Mochi 1 preview. A new SOTA in open-source video generation. Apache 2.0.

magnet:?xt=urn:btih:441da1af7a16bcaa4f556964f8028d7113d21cbb&dn=weights&tr=udp://tracker.opentrackr.org:1337/announce

[Chubby♨️ @kimmonismus]

Here we see GTA IV overlaid with AI generated graphics. I can very well imagine that in the near future we will generate graphics with AI, because it is cheaper and faster than if everything is created in advance in a graphics engine.

[hardmaru @hardmaru]

Exciting related work: “Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers”

New paper by @ChengleiSi @tatsu_hashimoto @Diyi_Yang @stanfordnlp, with help from 100+ NLP researchers❗

Paper: https://t.co/T3on9DMn1I
https://t.co/R3AJbzgSpo

[Jakob Foerster @j_foerst]

Scientific progress is one of humanity's most impressive and impactful intellectual achievements. We introduce The AI Scientist, the first AI to carry out end-to-end science, from ideation to implementation, data analysis, struggling w/ latex, reviewing and iterative improvement!

[hardmaru @hardmaru]

Flux is the new Stable Diffusion 🔥

https://t.co/iNP1A7GG2g

[Tris Warkentin @triswarkentin]

Gemma 2 - 2B is live!

The results are incredible for a small model -- better on LMSys Chatbot Arena than all GPT-3.5 models and besting Llama-2-70B-chat, which was very recently the biggest and best model! The pace of change is amazing.