Add Ollama backend + Qwen3 local inference support

- Extractor now supports two backends: ollama (local) and anthropic (cloud)
- Default is ollama with qwen3:14b (fits 16GB VRAM)
- Set num_ctx to 32768 for full-script processing
- Added --backend and --ollama-url CLI flags
- Added The Last Backup test script
- Tested: 12/12 scenes valid on dialogue_heavy, 12/13 on the_last_backup

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

This commit is contained in:

profit

2026-04-06 16:28:53 -07:00

co-authored by

Claude Opus 4.6

parent 87d0af0748

commit 74870f7c0d

5 changed files with 548 additions and 35 deletions

requirements.txt

View File

@@ -1,3 +1,4 @@
 anthropic>=0.40.0
 pydantic>=2.0.0
 python-dotenv>=1.0.0
 requests>=2.31.0