Show HN: Mordiem – OpenAI-compatible API with multi-network failover

Mordiem launched an OpenAI-compatible API that automatically fails over across multiple provider keys and networks when one is rate-limited or down, letting callers keep a single endpoint and API key. The service covers text, images, video, audio, and embeddings through one interface, lists per-token and per-image prices up front with no subscription, and bills users' own provider keys at cost. Developers point the OpenAI SDK at base_url https://api.mordiem.com/api/v1 with an mr_sk_ key and use models such as llama-3.3-70b.

Text, images, video, and audio Every modality through one interface. Point an OpenAI SDK at Mordiem and keep the same request shape. ChatImagesVideoSpeechEmbeddings Requests fail over across keys and providers when one is rate-limited or down. Callers keep a single endpoint. Per-token and per-image prices are listed up front. No subscription, and your own provider keys are billed at cost. Live catalog prices. Text is per million tokens. Images are per generation. | Model | Context | Input / 1M | Output / 1M | |---|---|---|---| | Loading catalog… | | | | OpenAI-compatible. Failover is automatic. python from openai import OpenAI client = OpenAI api key="mr sk ...", base url="https://api.mordiem.com/api/v1", One key. Failover is automatic. client.chat.completions.create model="llama-3.3-70b", messages= {"role": "user", "content": "Hello"} ,