d6eafe0989
- docs/README.md: add full documentation index (was empty) - docs/servers/ROCINANTE.md: create new server doc (RTX 4090, heavy inference) - docs/servers/SERENITY.md: fix CPU spec (dual X5690, not single); add reranker service - docs/architecture/SERVICES_DIRECTORY.md: fill in all service ports (was all dashes) - docs/architecture/ARCHITECTURE_OVERVIEW.md: mark all services as deployed - docs/planning/HOMELAB_EXPANSION_PLAN.md: mark all 6 phases complete
32 lines
796 B
Markdown
32 lines
796 B
Markdown
# Rocinante
|
|
|
|
Heavy AI inference server.
|
|
|
|
## Hardware
|
|
|
|
| Component | Details |
|
|
|-----------|---------|
|
|
| **IP** | 10.5.1.112 |
|
|
| **GPU** | NVIDIA RTX 4090 (24 GB VRAM) |
|
|
| **Role** | Heavy AI inference (Ollama) |
|
|
|
|
## Services
|
|
|
|
### Ollama
|
|
|
|
Ollama runs on port `11434` and hosts the large-model ("heavy") tier.
|
|
|
|
| Model | Use |
|
|
|-------|-----|
|
|
| `qwen3:latest` | Fast reasoning / general chat |
|
|
| `qwen2.5:32b` | Large-context tasks |
|
|
| `gemma3:27b` | Multimodal / vision |
|
|
|
|
> **LiteLLM tier:** Rocinante is the **"heavy"** tier in LiteLLM `config.yaml`.
|
|
> Heavy-tier requests from OpenWebUI and n8n route to `http://10.5.1.112:11434`.
|
|
|
|
## Notes
|
|
|
|
- No SSH config needed for routine use — Ollama API is accessed over LAN only
|
|
- Do not run multiple large VRAM allocations simultaneously on the 4090
|