Solution Blueprint Catalog#
The Solution Blueprint catalog ranges from simple demos and basic interfaces to fully-fledged agentic platforms and development environments. The catalog is constantly expanded to showcase the cutting edge innovations and to cover the established AI use cases.
Solution Blueprints are designed to run on AMD Instinct, Radeon, and EPYC. Browse the catalog below to find blueprints for your use case.
ROCm version
Solution Blueprints are tested with ROCm 7.2.3.
Blueprint |
Description |
AIMs |
Resources |
|---|---|---|---|
Decoupled RAG blueprint with an MCP Knowledge Server and an agentic UI client. |
GPT OSS 20B, Embedding |
2 GPU, 13 CPU, 272 GiB RAM |
|
AI-powered UI testing with an LLM agent that interprets specs, automates browser tests, and generates CI-ready test code. |
GPT OSS 120B |
1 GPU, 8 CPU, 72 GiB RAM |
|
Multi-agent translation workflow where LLM agents critique and refine outputs to improve translation quality. |
Llama 3.3 70B |
1 GPU, 5 CPU, 68 GiB RAM |
|
No-code visual platform to design, test, and deploy agentic AI workflows with language models and tools. |
Llama 3.3 70B |
1 GPU, 6 CPU, 68 GiB RAM |
|
Agentic documentation assistant that analyzes repositories and generates structured software architecture and component docs. |
Llama 3.3 70B |
1 GPU, 4 CPU, 64 GiB RAM |
|
Browser-based AI pair programmer with a local LLM for code completion, fixes, and interactive coding support. |
Qwen 3 32B, Qwen 2 5 Coder 7B |
3 GPU, 12 CPU, 256 GiB RAM |
|
Multimodal summarization service for text, PDF, DOCX, audio, and video content. |
Llama 3.3 70B, Whisper |
1 GPU, 5 CPU, 65 GiB RAM |
|
KYC onboarding workflow for fintech customer verification. |
Mistral Small 3 2 24B 2506 |
2 GPU, 4 CPU, 64 GiB RAM |
|
Financial analysis assistant combining stock data, technical indicators, and LLM reasoning for market insights. |
Llama 3.3 70B |
1 GPU, 5 CPU, 68 GiB RAM |
|
Sandbox chat interface to test prompts, evaluate outputs, and understand LLM behavior before production use. |
Llama 3.1 8B |
1 GPU, 5 CPU, 68 GiB RAM |
|
Prompt router that classifies requests and sends them to the best LLM endpoint using configurable routing rules. |
Llama 3.1 8B, Llama 3.3 70B, Embedding |
1 GPU, 4 CPU, 192 GiB RAM |
|
LiveKit-based voice consultation app with agent backend, web UI, and LLM support for medical assistance. |
Llama 3.3 70B, Qwen ASR |
2 GPU, 16 CPU, 132 GiB RAM |
|
MRI analysis tool with a Gradio interface and LLM-generated reports. |
GPT OSS 20B |
1 GPU, 5 CPU, 68 GiB RAM |
|
End-to-end pipeline that turns PDF documents into podcast-style audio using agentic orchestration, LLMs, and TTS. |
Llama 3.3 70B, Qwen TTS |
2 GPU, 16 CPU, 104 GiB RAM |
|
Multi-agent preventive healthcare outreach system that identifies screening candidates and drafts personalized emails. |
Llama 3.3 70B |
1 GPU, 5 CPU, 68 GiB RAM |
|
Multi-stage LLM engine that performs web research and produces structured, citation-backed technical reports. |
Llama 3.3 70B |
1 GPU, 2 CPU, 68 GiB RAM |
|
RAG application for chatting with documents using ChromaDB retrieval and LLM-based answers. |
Llama 3.3 70B, Embedding |
2 GPU, 11 CPU, 268 GiB RAM |
|
Real-time telecom voice assistant for customer support, account actions, and ticketing workflows. |
GPT OSS 120B, Mistral Small 3 2 24B 2506, Embedding, Qwen ASR, Qwen TTS |
5 GPU, 34 CPU, 421 GiB RAM |
Blueprint |
Description |
AIMs |
Resources |
|---|---|---|---|
Decoupled RAG blueprint with an MCP Knowledge Server and an agentic UI client. |
Qwen 3 Vl 8B, Embedding |
2 GPU, 13 CPU, 80 GiB RAM |
|
AI-powered UI testing with an LLM agent that interprets specs, automates browser tests, and generates CI-ready test code. |
Qwen 3 Vl 8B |
1 GPU, 8 CPU, 40 GiB RAM |
|
Multi-agent translation workflow where LLM agents critique and refine outputs to improve translation quality. |
Qwen 3 Vl 8B |
1 GPU, 5 CPU, 36 GiB RAM |
|
No-code visual platform to design, test, and deploy agentic AI workflows with language models and tools. |
Qwen 3 Vl 8B |
1 GPU, 6 CPU, 36 GiB RAM |
|
Multimodal summarization service for text, PDF, DOCX, audio, and video content. |
Qwen 3 Vl 8B, Whisper |
1 GPU, 5 CPU, 33 GiB RAM |
|
KYC onboarding workflow for fintech customer verification. |
Qwen 3 Vl 8B |
2 GPU, 4 CPU, 32 GiB RAM |
|
Financial analysis assistant combining stock data, technical indicators, and LLM reasoning for market insights. |
Qwen 3 Vl 8B |
1 GPU, 5 CPU, 36 GiB RAM |
|
Sandbox chat interface to test prompts, evaluate outputs, and understand LLM behavior before production use. |
Qwen 3 Vl 8B |
1 GPU, 5 CPU, 36 GiB RAM |
|
Prompt router that classifies requests and sends them to the best LLM endpoint using configurable routing rules. |
Llama 3.1 8B, Qwen 3 Vl 8B, Embedding |
1 GPU, 4 CPU, 32 GiB RAM |
|
LiveKit-based voice consultation app with agent backend, web UI, and LLM support for medical assistance. |
Qwen 3 Vl 8B, Qwen ASR |
2 GPU, 16 CPU, 100 GiB RAM |
|
MRI analysis tool with a Gradio interface and LLM-generated reports. |
Qwen 3 Vl 8B |
1 GPU, 5 CPU, 36 GiB RAM |
|
End-to-end pipeline that turns PDF documents into podcast-style audio using agentic orchestration, LLMs, and TTS. |
Qwen 3 Vl 8B, Qwen TTS |
2 GPU, 16 CPU, 72 GiB RAM |
|
Multi-agent preventive healthcare outreach system that identifies screening candidates and drafts personalized emails. |
Qwen 3 Vl 8B |
1 GPU, 5 CPU, 36 GiB RAM |
|
Multi-stage LLM engine that performs web research and produces structured, citation-backed technical reports. |
Qwen 3 Vl 8B |
1 GPU, 2 CPU, 36 GiB RAM |
|
RAG application for chatting with documents using ChromaDB retrieval and LLM-based answers. |
Qwen 3 Vl 8B, Embedding |
2 GPU, 11 CPU, 76 GiB RAM |
Blueprint |
Description |
AIMs |
Resources |
|---|---|---|---|
Agentic documentation assistant that analyzes repositories and generates structured software architecture and component docs. |
Llama 3.1 8B |
188 CPU, 128 GiB RAM |
|
Multimodal summarization service for text, PDF, DOCX, audio, and video content. |
Llama 3.1 8B, Whisper |
189 CPU, 129 GiB RAM |
|
Financial analysis assistant combining stock data, technical indicators, and LLM reasoning for market insights. |
Llama 3.1 8B |
189 CPU, 132 GiB RAM |
|
Sandbox chat interface to test prompts, evaluate outputs, and understand LLM behavior before production use. |
Llama 3.1 8B |
189 CPU, 132 GiB RAM |
|
Multi-stage LLM engine that performs web research and produces structured, citation-backed technical reports. |
Llama 3.1 8B |
189 CPU, 132 GiB RAM |
|
RAG application for chatting with documents using ChromaDB retrieval and LLM-based answers. |
Llama 3.1 8B, Embedding |
223 CPU, 172 GiB RAM |
Resources are example values, and can be reduced or increased depending on the user’s requirements.
Blueprint |
Description |
AIMs |
Resources |
|---|---|---|---|
A self-contained, fully open-source secure gateway for coding agents that runs entirely on one machine |
Qwen3-Coder-30B, AXIS, Lemonade SDK, GAIA |
Strix Halo Ryzen AI, 128 GiB RAM |