Here is a tutorial on training LLaSA (LLaMA-based TTS)... using GRPO to improve prosody, rhythm, and expressiveness in synthesized speech with TRL!show more

If you're using GRPO in TRL, you should really... switch to the new async trainer. In our benchmarks, it's ~2-4x faster 🔥show 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

Text-to-Speech is HERE 🎤 Create audiobooks, mini podcasts, and... your own TTS website with Manus!show more

with the launch of google's new IDE, it seems... appropriate to announce that my newest tutorial will be a fully web-based IDE (Cursor clone) using Next.js here is a sneak peek :)show more

🚀 Meet Qwen-TTS – now live via the Qwen... API ! Trained on millions of hours of speech, it delivers ultra-natural, expressive audio with smart prosody, pacing, and emotion. 🗣️ Supports 3 Chinese dialects: Beijing, Shanghai, Sichuan 🎙️ 7 bilingual voices: Cherry, Ethan, Chelsie, Serena, Dylan, Jada, Sunny 🔗 Learn more: 👉 Start using it via API:show more

This one explains why TTS is not just a... pronunciation problem i gave Cartesia and ElevenLabs the same sentence with bass, read, lead, bow, and a few other words that change pronunciation based on context Cartesia is currently #1 on the Artificial Analysis TTS leaderboard ElevenLabs is #5 This comparison makes the difference easier to hearshow more

Here is a quick tutorial on how to use... the new cl_netgraph in CS2show more

Scary: Patriots OT Will Campbell has been training nonstop... this offseason, using truck pushes to build strength and improve blocking. Will is going to be a PROBLEM this year 😈show more

we just released a new blog "Training a coding... agent using the OpenCode harness in remote HF sandboxes with TRL and OpenEnv" you can take a real coding agent (OpenCode), let it run its own tool loop against real coding problems, and train it with RL on the exact tokens it produced and every rollout runs in its own remote HF sandbox, so rollouts scale out beyond one machine the loop: - OpenCode owns its tool loop inside an OpenEnv sandbox - an in-sandbox proxy records the real token ids + logprobs, per turn - a hidden-test verifier scores the result, and that is the reward - TRL trains with AsyncGRPO, weights sync back to vLLM over NCCL blog + runnable example:show more

Tutorial: using GPT Image 2 and Seedance 2 to... create beautiful animated children's story. Here is how I created this. 👇show more

You can now do RL weight sync over HF... Buckets. > Async GRPO + LoRA in TRL > train on one Job, serve vLLM on others > sync just the small delta adapter through a Bucket. Trainer, inference and storage, all on Hugging Face infra. Beautifully optimised loop by @DirhoussssiAmine 👇 3h27 → 53 min for the same reward.show more

🚨 Marco Rubio just CRUSHED it. Even liberals don't... know what to say. My timeline is flooded with people raving about his speech in Munich that focused on defending the west, Christianity, and opposing mass migration. Based Marco is a force to be reckoned with, that's a FACT!show more

We found a new way to get language models... to reason. 🤯 No RL, no training, no verifiers, no prompting. ❌ With better sampling, base models can achieve single-shot reasoning on par with (or better than!) GRPO while avoiding its characteristic loss in generation diversity.show more

Tutorial on how to add an embroidered effect on... any object in illustrator with no plugin, using object > create mosaic + global editshow more

MiniMax Audio just leveled up with the new Speech-02... model! Turn any file or URL into lifelike audio instantly. Create audiobooks and podcasts effortlessly with up to 200k-character in a single input. Enjoy ultra-realistic TTS in 30+ languages with native flair, unlimited voice cloning, and sub-second streaming! Here are 4 examples:show more

WordVoice TTS ships what I've always wanted: a TTS... system with per word control you can let the system auto-pilot (based on CosyVoice3) or control every word with duration, loudness, pitch or tone works with cloned or pre-set voices ▶️ on spacesshow more