bail (Qwen)

Alibaba Bailian — Qwen-2.5/3 + Qwen-VL via the DashScope OpenAI-compatibility endpoint.

import { bail } from '@agentskit/adapters'

const adapter = bail({
  apiKey: process.env.DASHSCOPE_API_KEY!,
  model: 'qwen-max',
})

qwen is exported as an alias for the same factory.

#Options

Option	Type	Default
`apiKey`	`string`	required
`model`	`string`	`qwen-max`
`baseUrl`	`string`	`https://dashscope.aliyuncs.com/compatible-mode/v1`
`retry`	`RetryOptions`	inherited

#Capabilities

{ streaming: true, tools: true, multiModal: true, usage: true }. The compat endpoint mirrors the OpenAI chat-completions schema, so request/response shapes match openai({ baseUrl }).

#Why bail / Qwen

APAC latency + data residency.
Qwen-VL handles image / audio / video inputs.
Strong Chinese-first reasoning.

#Env

Var	Purpose
`DASHSCOPE_API_KEY`	API key

Providers overview · Choosing an adapter · openai

Explore nearby

Peer
Providers
25 native chat and embedder adapters, plus higher-order adapters that compose candidates. Separate from the 140-provider models catalog.
Peer
Choosing an adapter
Capability decision table and rules of thumb for picking a chat adapter.
Peer
Hosted chat adapters
17 managed-LLM adapters. Same contract; swap by changing one import.

✎ Edit this page on GitHub·Found a problem? Open an issue →·How to contribute →