Concepts
Providers & runtimes
How provider selection maps to managed runtime execution.
aex exposes one submission shape across supported providers:
| Provider | Selector |
|---|---|
| Anthropic | Providers.ANTHROPIC |
| DeepSeek | Providers.DEEPSEEK |
| OpenAI | Providers.OPENAI |
| Gemini | Providers.GEMINI |
| Mistral | Providers.MISTRAL |
| OpenRouter | Providers.OPENROUTER |
| Doubao | Providers.DOUBAO |
| Doubao China | Providers.DOUBAO_CN |
All submissions run on the managed runtime. The optional runtime option picks
a managed machine-size preset — use Sizes.* in TypeScript (e.g.
runtime: Sizes.SHARED_0_25X_1GB) or --runtime-size in the CLI. Omit it for
the default size; there is no alternative runtime backend to select.
Selection
TypeScript
import { Models, Providers } from "@aexhq/sdk";
await aex.run({
provider: Providers.OPENAI,
model: Models.GPT_4_1,
message: "Summarise the attached files.",
apiKeys: { openai: process.env.OPENAI_API_KEY! }
});CLI
aex run \
--api-key "$AEX_API_KEY" \
--provider openai \
--openai-api-key "$OPENAI_API_KEY" \
--model gpt-4.1 \
--prompt "Summarise the attached files." \
--followEvents, outputs, cleanup, and downloads use the same SDK and CLI surface for every provider. For the exact supported model list, use the generated provider/runtime capability matrix.