Extending Instella family of models with text-to-image model trained from scratch on AMD MI300X; fully open (dataset, training code, checkpoints and a detailed blog) to help reproducibility and pushing research forward. #AMD #AI #ML #Diffusion #MI300X #AMDGPU
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
Instella-T2I: Open-Source Text-to-Image with 1D Tokenizer and 32× Token Reduction on AMD GPUs
Explore Instella-T2I: AMD’s open-source text-to-image model, built on MI300X GPUs with novel tokenizer and LLM-based encoder for scalable image generation.
rocm.blogs.amd.com
July 16, 2025 at 5:34 AM
Extending Instella family of models with text-to-image model trained from scratch on AMD MI300X; fully open (dataset, training code, checkpoints and a detailed blog) to help reproducibility and pushing research forward. #AMD #AI #ML #Diffusion #MI300X #AMDGPU
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
Introducing Instella-Long model that we trained from scratch with up to 128K sequence length, fully open: checkpoints, code and dataset. And with competitive benchmark outperforming open-weights models on the long-context benchmark. #AMD #ML #AI #LLM #MI300
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
Introducing Instella-Long: A Fully Open Language Model with Long-Context Capability
Learn about Instella-Long: AMD’s open 3B language model supporting 128K context, trained on MI300X GPUs, outperforming peers on long-context benchmarks.
rocm.blogs.amd.com
June 12, 2025 at 6:19 AM
Speeding up your LLM inference with PARD (Parallel Draft Models), achieving up to 3.3x perf boost on llama 3 family of models!!!
www.amd.com/en/developer...
www.amd.com/en/developer...
April 11, 2025 at 4:40 PM
Speeding up your LLM inference with PARD (Parallel Draft Models), achieving up to 3.3x perf boost on llama 3 family of models!!!
www.amd.com/en/developer...
www.amd.com/en/developer...
AITER is all you need!!!
AI Tensor Engine for ROCm (AITER) a centralized repository for AMD high performance kernels that drastically boost inference and training performance.
rocm.blogs.amd.com/software-too...
AI Tensor Engine for ROCm (AITER) a centralized repository for AMD high performance kernels that drastically boost inference and training performance.
rocm.blogs.amd.com/software-too...
March 24, 2025 at 6:47 PM
AITER is all you need!!!
AI Tensor Engine for ROCm (AITER) a centralized repository for AMD high performance kernels that drastically boost inference and training performance.
rocm.blogs.amd.com/software-too...
AI Tensor Engine for ROCm (AITER) a centralized repository for AMD high performance kernels that drastically boost inference and training performance.
rocm.blogs.amd.com/software-too...
End-to-end multi-model RAG on AMD GPU!!!
Build an AI voice assistant with ROCm, @llamaindex.bsky.social, and RAG on AMD GPUs 🎙️🧠 Learn to create a pipeline that transcribes speech, uses RAG for responses, and converts text to speech. Step-by-step guide with @pytorch.org and Ollama:
https://rocm.docs.amd.com/projects/ai-developer-hub/en/
https://rocm.docs.amd.com/projects/ai-developer-hub/en/
March 21, 2025 at 9:33 PM
End-to-end multi-model RAG on AMD GPU!!!
Play Minecraft with AMD GPUs and AI PC!!!
A Mixture-of-Experts Agent for Open-World Minecraft that can complete various tasks in Minecraft automatically, with online RL training. It uses multi-modal information to identify and activate the most relevant expert for the target scenario.
A Mixture-of-Experts Agent for Open-World Minecraft that can complete various tasks in Minecraft automatically, with online RL training. It uses multi-modal information to identify and activate the most relevant expert for the target scenario.
www.amd.com
March 18, 2025 at 11:09 PM
Play Minecraft with AMD GPUs and AI PC!!!
A Mixture-of-Experts Agent for Open-World Minecraft that can complete various tasks in Minecraft automatically, with online RL training. It uses multi-modal information to identify and activate the most relevant expert for the target scenario.
A Mixture-of-Experts Agent for Open-World Minecraft that can complete various tasks in Minecraft automatically, with online RL training. It uses multi-modal information to identify and activate the most relevant expert for the target scenario.
Our newly optimized training Docker has been released. This update incorporates enhanced fine-tuning tools, improved multi-GPU performance, Megatron-LM support and many more. Give it a try, love to hear your feedbacks!!!
rocm.blogs.amd.com/software-too...
rocm.blogs.amd.com/software-too...
Optimized ROCm Docker for Distributed AI Training — ROCm Blogs
AMD updated Docker images incorporate torchtune finetuning, FP8 support, single node performance boost, bug fixes & updated benchmarking for stable, efficient distributed training
rocm.blogs.amd.com
March 14, 2025 at 4:16 AM
Our newly optimized training Docker has been released. This update incorporates enhanced fine-tuning tools, improved multi-GPU performance, Megatron-LM support and many more. Give it a try, love to hear your feedbacks!!!
rocm.blogs.amd.com/software-too...
rocm.blogs.amd.com/software-too...
Introducing AMD Intella3B, a fully open 3B LLM model trained from scratch on MI300X on 4 trillion tokens with competitive performance to similar size models. Everything is open!!! #AMD #ML #AI #LLM #AGI #MI300 #Instinct
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
Introducing Instella: New State-of-the-art Fully Open 3B Language Models — ROCm Blogs
AMD is excited to announce Instella, a family of fully open state-of-the-art 3-billion-parameter language models (LMs). , In this blog we explain how the Instella models were trained, and how to acces...
rocm.blogs.amd.com
March 6, 2025 at 7:28 AM
AMD ROCm user Meet Up @ Shack15, Thursday March 13. Try to attend, a packed event with all things AI on ROCm plus social networking and meeting the expert in the field. You don't want to miss!!! #AMD #ROCm #AI #LLM #ML #AMDGPU #Instinct
lu.ma/ow96nrie?tk=...
lu.ma/ow96nrie?tk=...
"AMD ROCm™ User Meet Up" @ Shack15 · Luma
Join us for the first "AMD ROCm™ User Meet Up" on March 13, 2025, from 5:00 PM to 9:00 PM at Shack15 in San Francisco's Ferry Building! Connect, learn,…
lu.ma
March 2, 2025 at 7:52 PM
Max-Achievable FLOPs Blog Part 2 is here!!! #AMD #ML #AI #GEMM #LLM
rocm.blogs.amd.com/software-too...
rocm.blogs.amd.com/software-too...
Measuring Max-Achievable FLOPs – Part 2 — ROCm Blogs
AMD measures Max-Achievable FLOPS through controlled benchmarking: real-world data patterns, thermally stable devices, and cold cache testing—revealing how actual performance differs from theoretical ...
rocm.blogs.amd.com
March 1, 2025 at 12:07 AM
Proud of the team, high quality text-to-video model that can run in your laptop!!! #AMD #ROCm #AI #ML #Diffusion #RyzenAI #Instinct #AMDGPU
www.amd.com/en/developer...
www.amd.com/en/developer...
www.amd.com
February 28, 2025 at 8:28 PM
Proud of the team, high quality text-to-video model that can run in your laptop!!! #AMD #ROCm #AI #ML #Diffusion #RyzenAI #Instinct #AMDGPU
www.amd.com/en/developer...
www.amd.com/en/developer...
Hands-on tutorial on ROCm!!! #AMD #ML #AI #ROCm #VLLM #TGI #PyTorch
rocm.docs.amd.com/projects/ai-...
rocm.docs.amd.com/projects/ai-...
Tutorials for AI developers — Tutorials for AI developers
The AI Developer Hub provides tutorials and guides for training, fine-tuning, and inference
rocm.docs.amd.com
February 12, 2025 at 6:03 PM
Running DeepSeek R1 on a single MI300X server!!!
techcommunity.microsoft.com/blog/AzureHi...
#AMD #ML #AI #LLM #Reasoning #Thinking #AGI
techcommunity.microsoft.com/blog/AzureHi...
#AMD #ML #AI #LLM #Reasoning #Thinking #AGI
Running DeepSeek-R1 on a single NDv5 MI300X VM | Microsoft Community Hub
Contributors: Davide Vanzo, Yuval Mazor, Jesse Lopez
DeepSeek-R1 is an open-weights reasoning model built on DeepSeek-V3, designed for conversational...
techcommunity.microsoft.com
February 5, 2025 at 3:55 AM
Running DeepSeek R1 on a single MI300X server!!!
techcommunity.microsoft.com/blog/AzureHi...
#AMD #ML #AI #LLM #Reasoning #Thinking #AGI
techcommunity.microsoft.com/blog/AzureHi...
#AMD #ML #AI #LLM #Reasoning #Thinking #AGI
Learn how to get the best performance out of MI300X, and take advantage of its memory capacity and bandwidth!!! #AMD #AI #ML #AGI #DeepLearning #HW #GPU
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
January 30, 2025 at 7:17 AM
Learn how to get the best performance out of MI300X, and take advantage of its memory capacity and bandwidth!!! #AMD #AI #ML #AGI #DeepLearning #HW #GPU
rocm.blogs.amd.com/artificial-i...
rocm.blogs.amd.com/artificial-i...
Interesting and scary, we need more research in arxiv.org/abs/2401.05566
January 13, 2024 at 9:20 PM
Interesting and scary, we need more research in arxiv.org/abs/2401.05566