Meta's Llama 4 Maverick, called through one OpenAI-compatible endpoint and paid from a prepaid wallet. 6 of our models cost less; prices are per million tokens, taken from the table billing charges from on 28 September 2026.
A thousand tokens each way costs about $0.000593 — roughly a page of text in and a page out. No reliable tool calling, so this model is hidden whenever live data is switched on.
Any OpenAI-compatible client works: change the base URL and the model id, keep everything else.
curl https://api.oraicle.me/v1/chat/completions \
-H "Authorization: Bearer $ORAICLE_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
"messages": [{"role": "user", "content": "Hello"}]
}'
The model id is meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8.
Full API docs →