Gemini 3.1 Flash Lite is Google's high-efficiency model optimized for high-volume, latency-sensitive use cases.
curl https://conduit-api.aurixel.ai/v1/chat/completions \
-H "Authorization: Bearer ck-YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gemini-3.1-flash-lite","messages":[{"role":"user","content":"Hello!"}]}'→ OpenAI-compatible. Add "stream": true (+ "stream_options":{"include_usage":true}) for SSE.
Reasoning effort: use model-name variants (-low / -agent). The reasoning_effort param is unreliable on Gemini.