Every AI provider, one key

Every AI provider on Aether API: OpenAI, Anthropic, Google, xAI, DeepSeek, Alibaba, Mistral and more. All of their models behind one OpenAI-compatible key, at least 30% below list price.

Every provider in the catalogue is reachable through one OpenAI-compatible key, with per-token pricing and at least 30% below the provider's own list price.

The providers

• OpenAI — GPT chat models, image generation, audio, and embeddings, reachable through the same key as every other model.
• Anthropic — Claude models for chat, long documents, and coding, priced below Anthropic's own list price.
• Google — Gemini models with very large context windows, for documents, code, and multimodal input.
• xAI — Grok models for chat and reasoning, on the same OpenAI-compatible endpoint as the rest.
• DeepSeek — DeepSeek chat and reasoning models at a fraction of the price of comparable frontier models.
• Alibaba — Qwen models for chat, code, and multimodal input, with strong context lengths.
• Perplexity — Perplexity models for search-grounded answers, on the same key and the same endpoint.
• Mistral — Mistral models for chat, code, and embeddings, in the open-weight tradition.
• Z.AI — Z.AI models for chat and reasoning, with GLM models for multilingual work.
• Moonshot — Kimi models for long context and coding, including the K2 code line.
• Minimax — Minimax chat and text-to-speech models, next to the rest of the catalogue.
• Black Forest Labs — FLUX image models for text-to-image and image editing, per image.
• Xiaomi — Xiaomi MiMo models for chat, added to the catalogue like every other provider.
• Runware — Runware image models, priced per image and reachable through the same key.
• Together — Together models for chat and open-weight work, on the shared endpoint.
• StepFun — StepFun models for chat, in the same catalogue and under the same key.

Related

Compare every model on the models page, or start with the quickstart.

Aether API — OpenAI-compatible access to 13+ providers through one API key.