Brev.dev reposted this
My Brev.dev Origin Story. Sadly the Arabic didn’t stick! I blame Nader Khalil
The Missing Google Colab Pro Tier. Fine-tune, train, or deploy. Use your own notebook, or one of ours. SSH too. CUDA, Python, Jupyter Lab, all set up.
External link for Brev.dev
San Francisco, CA, US
Chattanooga, TN, US
Brev.dev reposted this
My Brev.dev Origin Story. Sadly the Arabic didn’t stick! I blame Nader Khalil
Brev.dev reposted this
Metas new model Chameleon blew my mind 🤯 It's multi-modal, so text and images, but it does so using the same encoder! Unlike image gen models, it uses tokens not diffusion! To do this, it has to convert the text and images into a single sequence of tokens when processing. Then, a single transformer processes these mixed-modal tokens, eliminating the need for a different encoder per modality. If this works well, it might make for more scalable and efficient multi-modal models! I'll put a link to run it in the comments 🤙 Note: they made it "safety-aligned" and ripped out the ability to generate images, however, that code is still in there. If you are able to unlock it, I'll give you $1000
Brev.dev reposted this
Watch the demo, then run it yourself with Brev.dev
✨ #TensorRT and GeForce #RTX unlock ComfyUI SD superhero powers 🦸⚡ 🎥 Demo: https://nvda.ws/3wTg40b 📗 DIY notebook: https://nvda.ws/3XafSV4 ✨
Brev.dev reposted this
Co-Founder @ WeCracked | CS @ UM-Dearborn | Fullstack Engineer | WebDev Freelancer | Co-Project Coordinator for AIC | GDSC Dev Team Member | Intel Student Ambassador | CodePath TIP 102 | SAFA E-Board 🇵🇭
Hey LinkedIn 👋 A couple days ago, I attended the Anything But Wrappers: NVIDIA Llama3 NIM Edition "hackmeetup" hosted by Brev.dev and NVIDIA at the Solaris AI office in San Francisco. In 2023, scams targeting people aged 60 and older caused over $3.4 billion in losses, which is an 11% increase from 2022. The average victim lost $33,915, and 5,920 people lost more than $100,000. Now, imagine one of these targeted people was your grandma. Terrible right? Introducing, Granny Guard 👵💂 A white hat hacker tool designed to combat scammers by using reinforcement learning to train your grandma to identify them. My team, comprised of Joe Malatesta, Leon, Oscar Hong, Alexander Makhratchev, and Dennis Xing, used NVIDIA NIM and Brev to host a Llama3 model (enhanced with some prompt engineering!) that would interface with RetellAI's API services to provide real-time phone call conversations. Another AI agent would be used for sentiment analysis to detect if the Granny Guard scam was successful, and would notify you. Interested in a video demo? Check it out here: https://lnkd.in/eJWp_DWz Thank you Nader Khalil, Sarah Chieng, Michael Balint, and Julio C. Tapia for hosting such a fun event :) #brev #nvidia #nvidianim #llama3 #retellai #generativeai #llama #oss #aiagent
Brev.dev reposted this
We are teaming up with NVIDIA to throw a NIM hackathon NVIDIA NIMs are an easy way to run AI models anywhere Free GPUs, beer, and early access to NVIDIA's Llama3 NIM a week before it's released 🤙 Come learn from the brightest and hack on the most cutting edge AI software Speakers: Robert Nishihara CEO Anyscale Beyang Liu -- CTO Sourcegraph Michael Balint -- NIM NVIDIA Joseph Nelson -- CEO Roboflow Ishan Dhanani -- AI Lead Brev.dev Apply here -> https://lnkd.in/dsWwr-Pc
Brev.dev reposted this
I launched a Jupyter notebook on an NVIDIA H100 GPU running on Akash in just a few minutes using Brev.dev Brev is by far the simplest way to fine-tune and train ML workloads, and I am super excited that Akash is able to provide them with the much-needed high-density GPUs.
Today, Brev.dev is officially launching support for Akash GPUs. Brev is on a mission to create the easiest way to develop with the leading open-source AI models and deploy on GPUs in the cloud. Purpose-built for developers who prioritize speed and simplicity. Now, developers can 1-click deploy on NVIDIA H100s, A100s, and more at industry-leading pricing on the Akash Supercloud — directly from the Brev console. Get started with Brev: https://brev.dev/ Read the official coverage of the launch from Michael Nuñez of VentureBeat: https://lnkd.in/edKnQDFc
Run Llama 3 for inference on Brev! 🦙🤙
🦙🦙🦙 Optimize your Meta Llama 3 #inference with TensorRT-LLM …and how! ✨ in only 4 steps ✨ 📗➡️https://nvda.ws/44aUXSW AI at Meta
Brev.dev reposted this
🦙🦙🦙 Optimize your Meta Llama 3 #inference with TensorRT-LLM …and how! ✨ in only 4 steps ✨ 📗➡️https://nvda.ws/44aUXSW AI at Meta
Brev.dev reposted this
Today, Brev.dev is officially launching support for Akash GPUs. Brev is on a mission to create the easiest way to develop with the leading open-source AI models and deploy on GPUs in the cloud. Purpose-built for developers who prioritize speed and simplicity. Now, developers can 1-click deploy on NVIDIA H100s, A100s, and more at industry-leading pricing on the Akash Supercloud — directly from the Brev console. Get started with Brev: https://brev.dev/ Read the official coverage of the launch from Michael Nuñez of VentureBeat: https://lnkd.in/edKnQDFc
Fine-tune & deploy llama3 🤙
Fine-tune llama3 from scratch AND deploy it from the same jupyter notebook 🤙 Llama3 8B outperforms Llama2 70B in some use cases 🤯 This notebook guide uses SFT to fine-tune and vLLM for inference. Check out how incredible this SOTA model really is. Big s/o to Ishan Dhanani for getting this out in a timely manner 👇 https://lnkd.in/gxS78dW3