| Table of Contents |
|---|
Container Purpose | Option 1 (GPU) | Option 2 (CPU) |
|---|---|---|
Gateway (Linux) | 16 GB 4 vCPUs | 32 GB 8 vCPUs |
Embedder (Linux) | 16 GB 16GB GPU | None. Embedding happens on Gateway. |
LLM (Linux) | 16 GB 24 GB GPU | To be determined. |
Dashboard | t3a.large | t3a.large |
| Info |
|---|
|
AI model servers
The main hardware cost of the BusinessGPT deployment are the AI servers responsible for answering questions. These utilize GPUs.
...
Linux server Ubuntu 2CPU, 8GB RAM HDD/SSD with a R/W speed of at least 100MB/s.
GPU: CUDA 11.8+, Min 12GB 24GB RAM.
The disk size should be 30% larger than the original content base file size.
...