Gemma at 20k input tokens

[AnthonyRonning]

Our gemma model should be max 20k input tokens.

[AnthonyRonning]

@claude can you fix this? check where the model config is and set it to 20k

[claude]

Claude finished @AnthonyRonning's task —— View job • claude/issue-192-20250813-0420 • Create PR ➔

---

✅ Fixed Gemma 20k input tokens

Todo List:

• Search codebase for Gemma model configuration
• Locate and analyze current token limit settings
• Update Gemma model to 20k input tokens
• Run formatting and linting commands
• Test changes if possible
• Commit and push changes

🎯 Summary

Successfully updated both Gemma 3 27B model configurations in frontend/src/components/ModelSelector.tsx:

• google/gemma-3-27b-it: Changed from 70,000 → 20,000 tokens
• leon-se/gemma-3-27b-it-fp8-dynamic: Changed from 70,000 → 20,000 tokens

✅ Verification

• ✅ Code formatting passed
• ✅ Linting passed (only warnings, no errors)
• ✅ Build completed successfully
• ✅ Changes committed and pushed

---