Add Oracle service - LLM wrapper

First service for Vi's nervous system:
- Oracle service with NATS integration
- vLLM backend for Qwen3-32B
- GPTQ quantization support
- Thinking mode sampling configs

Simplified from Lyra's patterns, ready to test.

🦊
This commit is contained in:
Alex Kazaiev
2026-01-02 13:19:15 -06:00
parent e2d24a66f1
commit ee1cb5540a
8 changed files with 552 additions and 0 deletions

4
requirements.txt Normal file
View File

@@ -0,0 +1,4 @@
# Vi dependencies
nats-py>=2.6.0
vllm>=0.4.0
torch>=2.0.0