Join our community
Hardware Acceleration
Maximize performance with your iGPU
Data Security
Keep your information confidential.
OpenAI Compatibility
Use your preferred tools and libraries.
Liger Kernels Leap the CUDA Moat: A Case Study with Liger, LinkedIn's SOTA Training Kernels on AMD GPU
This guide shows the impact of Liger-Kernels Training Kernels on AMD MI300X. The build has been verified for ROCm 6.2.
Recent Blog Posts
By EmbeddedLLM Team • 8 mins
Nov 5, 2024
Liger Kernels Leap the CUDA Moat: A Case Study with Liger, LinkedIn's SOTA Training Kernels on AMD GPU
This guide shows the impact of Liger-Kernels Training Kernels on AMD MI300X. The build has been verified for ROCm 6.2.
By EmbeddedLLM Team • 5 mins
Oct 28, 2024
See the Power of Llama 3.2 Vision on AMD MI300X
This blog post shows you how to run Meta's powerful Llama 3.2-90B-Vision-Instruct model on an AMD MI300X GPU using vLLM. We provide the Docker commands, code snippets, and a video demo to help you get started with image-based prompts and experience impressive performance
By EmbeddedLLM Team • 8 mins
Oct 11, 2024
How to Build vLLM on MI300X from Source
This guide walks you through the process of building vLLM from source on AMD MI300X. The build has been verified for ROCm 6.2.
By EmbeddedLLM Team • 7 mins
Oct 27, 2023
High throughput LLM inference with vLLM and AMD: Achieving LLM inference parity with Nvidia
EmbeddedLLM has ported vLLM to ROCm 5.6, and we are excited to report that LLM inference has achieved parity with Nvidia A100 using AMD MI210.
Answers to your questions
Why using Embedded LLM platform does not require coding experience?
Using our platform is as simple as using a to-do list. We provide a highly intuitive and user-friendly interface that allows you to prototype and develop your own LLM pipeline swiftly and seamlessly. We also provide prompt templates to help you get started.
How can you ensure privacy and confidentiality of your LLM platform?
Our platform and services are bundled into our LLM appliance where you will have full control and ownership of the data and the LLM. Your data will always remain on-premise. Our appliance comes with access control function where administrators can set security measures based on organization’s policies.
How can I integrate your LLMs into my existing software infrastructure?
We offer API and SDK to interface with ELLM Appliances.
How is your Embedded LLM platform different from ChatGPT?
From our extensive experience advising clients on AI workflow automation, we found that there are 4 types of workload in a typical enterprise: (1) real-time ad-hoc, (2) real-time recurring, (3) batch ad-hoc, and (4) batch recurring. Current ChatGPT interface only addresses the realtime ad-hoc workflow. With our workflow autopilot interface you can breakdown a task into a series of steps just like a project manager and schedule the tasks to be done at a certain time or according to a schedule.