Our new flagship Olmo 3 model family empowers the open source community with not only state-of-the-art open models, but the entire model flow and full traceability back to training data.

...

Quick thoughts on another fantastic open model from a rapidly rising Chinese lab.

Contribute to apple/ml-simplefold development by creating an account on GitHub.

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024) - hiyouga/LLaMA-Factory

A Blog post by Stefano Fiorucci on Hugging Face

The GPT-5 launch was uh, rough. A lot went wrong here, and I want to talk about what really happened...Thank you Kilo Code for sponsoring! Check them out at:...

Hobbyist training AI on Victorian texts gets an unexpected history lesson from his own creation.

What’s the strongest model I can train on my MacBook Pro in five minutes? I’ll give the answer upfront: the best 5-minute model I could train was a ~1.8M-param…

The long promised OpenAI open weight models are here, and they are very impressive. They’re available under proper open source licenses—Apache 2.0—and come in two sizes, 120B and 20B. OpenAI’s …

A paper from Anthropic describing persona vectors and their applications to monitoring and controlling model behavior

The Moonshot AI team synthesized thousands of tools, agents, users, and sessions to build a library of training data.

A practical handbook for engineers building, optimizing, scaling and operating LLM inference systems in production.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

In this video, I break down Proximal Policy Optimization (PPO) from first principles, without assuming prior knowledge of Reinforcement Learning. By the end, you’ll understand the core RL building blocks that led to PPO, including: 🔵 Policy Gradient 🔵 Actor-Critic Models 🔵 The Value Function 🔵 The Generalized Advantage Estimate In the LLM world, PPO was used to train reasoning models like OpenAI's o1/o3, and presumably Claude 3.7, Grok 3, etc. It’s the backbone of Reinforcement Learning with Human Feedback (RLHF) -- which helps align AI models with human preferences and Reinforcement Learning with Verifiable Rewards (RLVR), which gives LLMs reasoning abilities. Papers: - PPO paper: https://arxiv.org/pdf/1707.06347 - GAE paper: https://arxiv.org/pdf/1506.02438 - TRPO paper: https://arxiv.org/pdf/1502.05477 Well-written blogposts: - https://danieltakeshi.github.io/2017/04/02/notes-on-the-generalized-advantage-estimation-paper/ - https://huggingface.co/blog/NormalUhr/rlhf-pipeline - https://iclr-blog-track.github.io/2022/03/25/ppo-implementation-details/ Implementations: - (Original) OpenAI Baseslines: https://github.com/openai/baselines/blob/ea25b9e8b234e6ee1bca43083f8f3cf974143998/baselines/ppo2 - Hugging Face: https://github.com/huggingface/trl/blob/main/trl/trainer/ppo_trainer.py - Hugging Face docs: https://huggingface.co/docs/trl/main/en/ppo_trainer Mother of all RL books (Barto & Sutton): http://incompleteideas.net/book/RLbook2020.pdf 00:00 Intro 01:21 RL for LLMs 05:53 Policy Gradient 09:23 The Value Function 12:14 Generalized Advantage Estimate 17:17 End-to-end Training Algorithm 18:23 Importance Sampling 20:02 PPO Clipping 21:36 Outro Special thanks to Anish Tondwalkar for discussing some of these concepts with me. Note: At 21:10, A_t should have been inside the min. Thanks @t.w.7065 for catching this.

Normalizing Flows (NFs) are likelihood-based models for continuous inputs. They have demonstrated promising results on both density…

35% off our evals course: https://bit.ly/evals-ai Vincent introduces Marimo, a reactive notebook environment. He walks us through the features of Marimo, including interactive and reactive charts and widget integration. Vincent demonstrates how you can use these components to build annotation apps for evals. Vincent also highlights differences between Marimo and traditional Jupyter Notebooks. Links: 1. Repo w/notebook: https://github.com/koaning/molabel 2. Vincen'ts drawing pad: https://www.amazon.com/Inspiroy-H640P-Graphics-Battery-Free-Sensitivity/dp/B075T6MTJX 3. Vincent's sites: https://koaning.io , and https://calmcode.io/ 00:00 Introduction to Data Science Journey 00:27 Exploring the Chick Weight Dataset 00:57 Interactive Data Analysis with Marimo 02:04 Importance of Looking at Data 03:32 Advanced Data Visualization Techniques 05:14 Introduction to Marimo's Unique Features 06:44 Reactive Programming in Marmo 12:50 AI Integration and Custom Rules 15:30 Marimo's Storage and Export Options 27:16 Advanced Visualization and Annotation 37:10 Introduction to Any Widget 37:45 Building Custom Widgets 38:56 Showcasing the Scatter Widget 40:29 Defining Widgets with Any Widget 45:58 Annotation Widgets and Their Uses 52:14 Exploring More Widget Capabilities 01:01:32 Marimo's App Mode and Deployment 01:03:37 Final Thoughts and Future Directions 01:04:45 Q&A and Closing Remarks

To create reliable agents, AI companies had to go beyond predicting the next token.

Get started with 10Web and their AI Website Builder API: https://10web.io/website-builder-api/?utm_source=YouTube&utm_medium=Influencer&utm_campaign=TechWithTim Today, you'll learn how to fine-tune LLMs in Python for use in Ollama. I'll walk you through it step by step, give you all the code and show you how to test it out. DevLaunch is my mentorship program where I personally help developers go beyond tutorials, build real-world projects, and actually land jobs. No fluff. Just real accountability, proven strategies, and hands-on guidance. Learn more here - https://training.devlaunch.us/tim ⏳ Timestamps ⏳ 00:00 | What is Fine-Tuning? 02:25 | Gathering Data 05:52 | Google Collab Setup 09:17 | Fine-Tuning with Unsloth 16:58 | Model Setup in Ollama 🎞 Video Resources 🎞 Code in this video: https://drive.google.com/drive/folders/1p4ZilsJsdxB5lH6ZBMdIEJBt0WVUMsDq?usp=sharing Notebook Google Collab: https://colab.research.google.com/drive/1NsRGmHVupulRzsq9iUTx8V8WgTSpO_04?usp=sharing Hashtags #Python #Ollama #LLM

Try out Warp 2.0 now, the current rank #1 AI on Terminal Bench, outperforming Claude Code: https://go.warp.dev/bycloud You can also use code "BYCLOUD" to get Warp Pro for 1 month free. (limited for 1,000 redemptions) My Newsletter https://mail.bycloud.ai/ my project: find, discover & explain AI research semantically https://findmypapers.ai/ My Patreon (get bundle access for my newsletter & findmypapers) https://www.patreon.com/c/bycloud Training language models to follow instructions with human feedback [Paper] https://arxiv.org/abs/2203.02155 DeepSeek-R1 (Aha Moment) [Paper] https://arxiv.org/abs/2501.12948 Understanding R1-Zero-Like Training: A Critical Perspective [Paper] https://arxiv.org/pdf/2503.20783 Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model? [Paper] https://arxiv.org/abs/2504.13837 Reinforcement Learning Finetunes Small Subnetworks in Large Language Models [Paper] https://arxiv.org/abs/2505.11711 Spurious Rewards: Rethinking Training Signals in RLVR [Paper] https://arxiv.org/abs/2506.10947 Try out my new fav place to learn how to code https://scrimba.com/?via=bycloudAI This video is supported by the kind Patrons & YouTube Members: 🙏Nous Research, Chris LeDoux, Ben Shaener, DX Research Group, Poof N' Inu, Andrew Lescelius, Deagan, Robert Zawiasa, Ryszard Warzocha, Tobe2d, Louis Muk, Akkusativ, Kevin Tai, Mark Buckler, NO U, Tony Jimenez, Ângelo Fonseca, jiye, Anushka, Asad Dhamani, Binnie Yiu, Calvin Yan, Clayton Ford, Diego Silva, Etrotta, Gonzalo Fidalgo, Handenon, Hector, Jake Disco very, Michael Brenner, Nilly K, OlegWock, Daddy Wen, Shuhong Chen, Sid_Cipher, Stefan Lorenz, Sup, tantan assawade, Thipok Tham, Thomas Di Martino, Thomas Lin, Richárd Nagyfi, Paperboy, mika, Leo, Berhane-Meskel, Kadhai Pesalam, mayssam, Bill Mangrum, nyaa, Toru Mon [Discord] https://discord.gg/NhJZGtH [Twitter] https://twitter.com/bycloudai [Patreon] https://www.patreon.com/bycloud [Business Inquiries] bycloud@smoothmedia.co [Profile & Banner Art] https://twitter.com/pygm7 [Video Editor] @Booga04 [Ko-fi] https://ko-fi.com/bycloudai

That gameboy blender animation...took 6 hours to render 😅. Anyway, had a ton of fun coding this up and finally getting back to some proper ML. I've been thi...

5 years ago, nobody would have guessed that scaling up LLMs would as successful as they are. This belief, in part, was due to the fact that all known statistical learning theory predicted that massively oversized models should overfit, and hence perform worse than smaller models. Yet the undeniable fact is that modern LLMs do possess models of the world that allow them to generalize beyond their training data. Why do larger models generalize better than smaller models? Why does training a model to predict internet text cause it to develop world models? Come deep dive into the inner working of neural network training to understand why scaling LLMs works so damn well. Want to see more videos like this in the future? Support me on Ko-fi https://ko-fi.com/algorithmicsimplicity Papers referenced: Double Descent: https://arxiv.org/abs/1812.11118 The Lottery Ticket Hypothesis: https://arxiv.org/abs/1803.03635 My previous videos on Autoregressive Transformers: Auto-regression (and diffusion): https://youtu.be/zc5NTeJbk-k Transformers: https://youtu.be/kWLed8o5M2Y

Implementation of the Aurora model for Earth system forecasting - microsoft/aurora

In this article, we carry out handwritten text recognition using OCR. We fine tune the TrOCR model on the GNHK dataset.

Anthropic publish most of the system prompts for their chat models as part of their release notes. They recently shared the new prompts for both Claude Opus 4 and Claude …

Once errors creep into the AI knowledge base, they can be very hard to get out.

The simplest, fastest repository for training/finetuning small-sized VLMs. - huggingface/nanoVLM

Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation - MoonshotAI/Kimi-Audio

Reinforcement Learning has helped train neural networks to win games, drive cars and even get ChatGPT to sound more human when it responds to your prompt. Th...