feat: Multi-modal support with automatic model fallback

Browse Source

- Add MultiModalManager with capability detection for vision/audio/tools
- Define fallback chains: vision (llama3.2:3b -> llava:7b -> moondream)
                       tools (llama3.1:8b-instruct -> qwen2.5:7b)
- Update CascadeRouter to detect content type and select appropriate models
- Add model pulling with automatic fallback in agent creation
- Update providers.yaml with multi-modal model configurations
- Update OllamaAdapter to use model resolution with vision support

Tests: All 96 infrastructure tests pass

...

This commit is contained in:

Alexander Payne

2026-02-26 22:29:44 -05:00

parent a85661274c

commit 72a58f1f49

8 changed files with 990 additions and 14 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

BIN

data/scripture.db-shm Normal file

Binary file not shown.