feat: add native Anthropic auxiliary vision

This commit is contained in:
teknium1 2026-03-14 21:14:20 -07:00
parent dc11b86e4b
commit db362dbd4c
7 changed files with 386 additions and 30 deletions

View file

@ -3,7 +3,8 @@
Vision Tools Module
This module provides vision analysis tools that work with image URLs.
Uses Gemini 3 Flash Preview via OpenRouter API for intelligent image understanding.
Uses the centralized auxiliary vision router, which can select OpenRouter,
Nous, Codex, native Anthropic, or a custom OpenAI-compatible endpoint.
Available tools:
- vision_analyze_tool: Analyze images from URLs with custom prompts
@ -409,7 +410,7 @@ if __name__ == "__main__":
if not api_available:
print("❌ No auxiliary vision model available")
print("Set OPENROUTER_API_KEY or configure Nous Portal to enable vision tools.")
print("Configure a supported multimodal backend (OpenRouter, Nous, Codex, Anthropic, or a custom OpenAI-compatible endpoint).")
exit(1)
else:
print("✅ Vision model available")