Lots of useful new features for large scale distributed training with #PyTorch on #Composer's new release.

How many hours of your life do you lose to debugging CUDA OOM errors? Automatic gradient accumulation can save you valuable time: mosaicml.com/blog/farewell-…

I'll be at MLSys on Tuesday 8/30 - would love to connect and meet up, DM me!

Dream team at . 🚀

How much do you think it costs to train a GPT-3 quality model from scratch?

Our open source Composer library has everything to improve the ML model training experience. Today we're dropping our 0.10 release: @cometml integration, auto-eval batch size selection, API improvements, and streaming #datasets. mosaicml.com/blog/composer-…

Excited and proud to share what we've been working on! 👇🏽

Folks at Amazon Web Services know their stuff 👇🏽

Our VP of Engineering Hagay Lupesko gave a great talk at last week's #PyTorchConference on how easy it is to speed up #ML model #training with our Composer tool.⚡️ Couldn't make it to New Orleans? Catch the recap here! mosaicml.com/blog/superchar…

Been hinting at this blog for a while and it's finally here! The Streaming team . has built an open source library (`mosaicml-streaming`) for efficiently loading training data from object stores like S3, GCS, OCI, and more. mosaicml.com/blog/mosaicml-…

Think it’s too hard—or too expensive—to train your own GPT or diffusion models from scratch? Think again. We built the MosaicML platform to tackle the challenges of training large models and unleash the power of #generativeAI. Learn more: mosaicml.com/blog/train-cus…

A quintessential example of the importance of data privacy and why companies should build and use their own models. Demonstrates the significance of what Naveen Rao, Jonathan Frankle, and Hagay Lupesko are building at .. techradar.com/news/samsung-w…

Woo hoo! 🙌What an honor to make the Forbes AI 50 List. MosaicML empowers you build your own #GenerativeAI. Train, finetune, and deploy your custom #LLM today: mosaicml.com

Tune in to hear from our CTO Hanlin Tang !

The current prevailing approach for developers is to use proprietary LLMs through APIs. Hagay Lupesko . on why factors such as domain specificity, security, privacy, regulations, IP protection & control, will prompt organizations to opt to invest in their own custom LLMs

🚨 A few months ago we announced that you can train Stable Diffusion from scratch for less than $125k using the MosaicML platform. A major price drop is coming...and we have the training run to back it up. Stay tuned for a major announcement this week!

Jonathan Frankle Landan Seguin Now, with ., it's absurdly easy. Here's the loss curve from the week. I launched the runs and didn't have to mess with the ongoing runs once, including when things broke during my Saturday brunch with Vivek Myers. It crashed, identified failure, restarted by itself 🤯

📣Announcing MosaicML Inference 📣 Ever wanted a text or image generation API that doesn’t make you send data to a third party? Or a cheaper solution than paying by the token? Or an easy way to get a trained model into production? We can help with that. 🧵

🎧 Our VP of Engineering Hagay Lupesko joins the Artificial Intelligence podcast to share insights on how our #GenAI platform enables #LLM training at scale. Give a listen! open.spotify.com/episode/5rcmHv…

🆕💡🎧 DBRX Unpacked - featuring Hagay Lupesko @Databricks: 🔥 Open LLM bridging quality & cost for AI apps ⚡️ Mixture-of-experts architecture for efficiency ⚙️ Optimized for enterprise needs: code gen, long context thedataexchange.media/dbrx-a-leap-fo…