Text, Vision, Thinking model from qwen. One provider serves it on AgentMarket, so every request routes to AIHubMix. Rates are per million tokens, in USD or in Units (100,000 Units = $1), drawn from one wallet balance that works across every provider.
The API is OpenAI-compatible. Swap the base URL and most SDKs work unchanged. Only the model slug differs between models.
curl https://router.fluxapay.xyz/llm/v1/chat/completions \
-H "Authorization: Bearer $MONETIZE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"qwen/qwen3.5-397b-a17b","messages":[{"role":"user","content":"Hello"}]}'Create an API key at /keys and export it as $MONETIZE_API_KEY. No provider in the path, so this routes to the cheapest one serving the model on chat completions — today AIHubMix. To pin one, add "provider": "aihubmix" to the request body.
AIHubMix is the only provider carrying this model, so there is no routing choice to make. Rates below are what you pay.
Prompt size counts every input token: new, cached and written to cache. Above a tier's size, all of the call's tokens are billed at that tier.