GPT-6 Astra test 4/n Prompt: "Implement pen spinning with... a dexterous hand. Use Isaac Lab for RL training, use the Sharpa hand, and create the pen mesh yourself. Give me a trained RL policy and a visualization video. You are free to search the web and download papers or anything else you need." After running autonomously for a day and a half (including training the policy), this is what we got: Credit: my student Chengyang Lishow more

New research from Databricks: LLMs Can Learn to Reason... via Off-Policy RL Optimal Advantage-based Policy Optimization with Lagged Inference policy (OAPL) shows you don’t need strict on-policy training to improve reasoning. It matches or beats Group Relative Policy Optimization (GRPO), stays stable with large policy lag, and uses ~3× fewer training generations. For Databricks customers, it’s a simpler, practical, and equally powerful approach to RL that Databricks is pioneering internally — and bringing directly to Databricks customers, so enterprises can improve agents using the same methods we use for our in-house agents, without complex infrastructure changes.show more

Introducing RL Environment Creator Skill Now any one can... create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.show more

Player stats UI animation with GPT Image 2 and... MiniMax H3. This model is amazing for UI animations. Give GPT Image 2 your character image and ask it to create a stats UI. Then use that UI as a reference for MiniMax H3. You can check the prompt in the replies.show more

This is my first "Holy Shit" moment with GPT-6... Astra I asked GPT-6 Astra to create a ROBLOX game, and the results are stunning! Who needs to make games when it's this easy to make them in the click of a button? AI Rules! I'm super excited to see what GPT-6 Astra is capable of!show more

As a supporter of the open weights ecosystem, we're... proud to be a post-training partner for NVIDIA Nemotron. We post-train Nemotron models for customer use cases, de-risk mainline RL runs on our AC2 platform and training stack, and contribute aggregate workload statistics for inference benchmarking. This is how open models get better, and we're excited to keep working closely with NVIDIA AI.show more

GPT-6 Astra + MiniMax H3 A workflow I've been... experimenting with: First, used GPT-6 Astra with Blender to create the base video as a simple 3D previs. Then took that video into MiniMax H3 and used a character reference + prompt to replace the placeholder object, populate the environment and transform the whole scene into a polished 2D anime sequence. The useful part is that Blender already solves the camera path, timing, scale, blocking and environment geometry. So instead of asking the video model to figure out everything from scratch, you give it a moving blueprint and let it focus on the character, animation and final look. You can find the MiniMax H3 prompt below. Used a premade model from sketchfab for this experiment: "Day 5. Final:The Western Saloon" ( by Olyona Kuznetsova is licensed under Creative Commons Attribution.show more

Create the viral “girlfriend opens the door” video with... GPT Image 2 + local MiniMax H3. 1️⃣ Use GenVizu Canvas Mode to generate the character and scene reference images. 2️⃣ Use my ComfyUI node to turn a simple prompt into a structured 6-part H3 prompt. 3️⃣ Wait & Enjoy! ✨show more

This astronaut on an open lunar rover is a... useful test case for **GPT-6 Astra × V2Fun** decomposition. The character silhouette, props or mechanics, materials, and pose relationships all need to match the prompt. Astra can reason about the structure and plan the components, while V2FUN handles the harder geometry and surface details. Fewer low-level tokens, faster iteration, cleaner assets. #V2Fun #GPT6 #Astra #AI3D #3DAssetshow more

Chroma keyable Daisy dance, free to use for whatever... you want with no credit needed! Will post a download for transparent source frames and the HD versions of the above animations as a reply a little later!show more

What a FUCKING JOKE these refs are. For anyone... who hasn’t seen this (where it HITS THE PSG PLAYER FIRST) this is an intentional hand ball by Miley and therefore a pen. Embarrassing.show more

RL is painfully slow 😭 — bottlenecked by super-long... CoT rollout. 🔭 Sparse attention should help, but naive sparse rollout hits a brutal efficiency–stability tradeoff: A tedious trial-and-error sparsity sweep for each dense policy is required before an actual RL run. 🐤Sparrow chirps no more pain! Introduce Sparrow: Sparse Rollout for stable and efficient long-context RL. Sparrow finds that: 💡As long as we keep the tail distribution mismatch throughout the sparse rollout above a critical threshold, the RL training will be stable. 💡Even cooler! Through comprehensive control studies of Qwen3-1.7B, 4B, 8B thinking models RL with 40K rollout max length, the critical threshold stays constant across model sizes. 💡Sparrow then finds the optimal dynamic sparse schedule to reach the threshold with minimal cost. 💡Sparrow's findings are empirically validated to generalize in Qwen3-14B, and hold on both Math and Coding RL. 🐤Sparrow empirically helps achieve 2.2× / 2.4× / 2.0× rollout speedup on Qwen3 1.7B / 4B / 8B thinking models, while keeping training stability over extended RL steps. We release the 🐤bird in the following formats. [1/n] Paper: Code: Blog:show more

I found a site you copy prompts for your... motion websites design Here: Copy the prompt and paste on lovable or any ai you use to create websites You are welcomeshow more

INTRODUCING: GPT 6-ASTRA for marketing just kidding. i'm giving... this away for free, runs locally on your device with one sentence you take one winning ad and make 50+ variations of it different faces, hooks, and angles for your business run the same campaign budget across all 50 instead of betting it on 1 video and calling it a flop download it here for free: (this video was made by gpt astra as well btw, prompt in the comments)show more

Yeaah with Kartel.ai and Jesse Wellens ! So training... Lora for character on Flux and also for Wan 2.1 14B ( using dataset video ) make a Huge difference. So here you are seeing a volumetric capture of Jesse Wellens that we did in at the spatial studio of Kartel.ai We use a a World Labs gaussian splatting to create a simple environment that we put with the volumetric capture in our own webgl viewer ( out soon for everyone ). After that, We use after this output in a ComfyUI workflow with Wan Fun control ( a bit similar Vace but its working for Wan 14B) with the double loras , the first to generate the first frame with flux, and the second one to guide the generation of Wan 2.1 Control. So we keep very good consistency of character and position, and creating amazing worlds :) ! Hope you will find it cool !show more

You can now post-train a model inside your existing... production harness with our platform, AC2. A production harness is a whole engineered system around the LLM, with its own context management, tools, sandboxing, and control flows. Porting that into a new training runtime can be expensive and could introduce train-test mismatch, where the policy is optimized against a simulated harness and then struggles in production. All you need to do is swap out the harness’ LLM response endpoint to one provided by AC2, and expose a lightweight protocol for AC2 to initiate and grade rollouts; the trainer handles the rest.show more

Fusion is the most efficient frontier harness for GPT... 6 Astra and Claude Fable 5.1. You can also configure everything about it: • Base model: Fable, Astra, Sol, or Opus • Sidekick model: SWE (Free), Luna, Sol, GLM (Free) • Speed: Normal or Fast ⚡ • Reasoning level It retains Claude Fable 5.1 and GPT-6 Astra performance while reducing costs and is the first time a multi-model coding agent has been included on the Artificial Analysis Coding Agent Index! Importantly: it's also fun to use :)show more

Code Interpreter in ChatGPT is incredible! Took me 5... mins to make this game. You can make your own game assets with any AI generator and then ask GPT-4 with Code Interpreter to write code. If you have any problems you can ask it to fix the errors. 1. Write this prompt: "write p5.js code for Asteroids where you control a spaceship with the mouse and shoot asteroids with the left click of the mouse. If your spaceship collides with an asteroid, you lose. If you shoot down all asteroids, you win! I want to use my own textures for the spaceship and for asteroids." 2. Go to Openprocessing website create and save sketch (you'll need to save it before uploading any texture files). Copy paste code from GPT-4 3. Generate texture files and remove backgrounds, for example in Clip Drop 4. Replace names of files with your filenames 5. Run the program 6. If something doesn't work ask GPT-4 to fix it (you can copy an error and paste in GPT-4) like you would ask a human programmer 7. To learn a bit of programming write these prompts to GPT-4: "Act as my programming teacher. Tell me an algorithm of Asteroids game in detail and make names of functions and explain what each of these functions will do. Don't write the code just yet." and then " Can you describe the algorithm overall for a 10-year-old child"show more