ai-agents-gemini-image-generat-calling-v1-images-generations-with-vertex-ai-gemin-7d9f71c3

THEN For Gemini image generation models, use /v1/chat/completions with modalities: ['text', 'image'] instead of /v1/images/generations. This routes through the correct :generateContent endpoint internally. Once LiteLLM deploys the fix, /v1/images/generations will work directly.

# Workaround: Use chat completions endpoint instead of images endpoint
curl -X POST http://localhost:4000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "vertex_ai/gemini-2.5-flash-image-preview",
    "messages": [{"role": "user", "content": "A beautiful sunset over mountains"}],
    "modalities": ["text", "image"]
  }'