alibaba

Documentation

Overview

Package alibaba wraps Alibaba Cloud's DashScope model platform, which hosts the Qwen / Tongyi family and several other Alibaba models.

DashScope exposes two surfaces:

• the native /api/v1/services/aigc/text-generation/generation endpoint with a DashScope-specific JSON shape (not used here);
• the /compatible-mode/v1 path which speaks the OpenAI chat-completions / embeddings spec.

This package uses the compatible-mode endpoint to route through the openai provider facade. DashScope-specific knobs (enable_thinking, enable_search, web search citations, etc.) use the namespaced OpenAI request extension.

See https://help.aliyun.com/zh/model-studio/ for the docs.

Index

• Constants
• type Chat
  • func NewChat(config ChatConfig) (*Chat, error)
• type ChatConfig
  • func (c ChatConfig) Validate() error
• type EmbeddingModel
  • func NewEmbeddingModel(config EmbeddingModelConfig) (*EmbeddingModel, error)
• type EmbeddingModelConfig
  • func (e EmbeddingModelConfig) Validate() error

Constants

const (
	OpenAIRequestExtensionKey     = "alibaba/openai_request"
	OpenAIResponseExtensionKey    = "alibaba/openai_response"
	OpenAIStreamChunkExtensionKey = "alibaba/openai_stream_chunk"
)

Namespacing preserves provider-specific data without promoting it into the shared Core protocol or colliding with another provider.

const (
	// BaseURLChina is the domestic DashScope OpenAI-compat endpoint.
	BaseURLChina = "https://dashscope.aliyuncs.com/compatible-mode/v1"

	// BaseURLIntl is the Singapore DashScope OpenAI-compat endpoint —
	// required for international (non-mainland-China) users.
	BaseURLIntl = "https://dashscope-intl.aliyuncs.com/compatible-mode/v1"
)

const (
	ModelQwen37Max   = "qwen3.7-max"
	ModelQwen37Plus  = "qwen3.7-plus"
	ModelQwen36Plus  = "qwen3.6-plus"
	ModelQwen36Flash = "qwen3.6-flash"

	// ModelQwen3CoderPlus (qwen3-coder-plus) targets code generation
	// / completion / repair.
	ModelQwen3CoderPlus = "qwen3-coder-plus"
)

Current chat model ids. The Qwen family is versioned aggressively; the constants below intentionally name rolling production aliases. See https://help.aliyun.com/zh/model-studio/getting-started/models for the live catalog.

const (
	// ModelEmbeddingV4 (text-embedding-v4) is the current general-purpose
	// embedding model.
	ModelEmbeddingV4 = "text-embedding-v4"
)

Embedding model ids.

const (
	Provider = "Alibaba"
)

Provider is the stable backend name for host-side attribution.

Types

type Chat

type Chat = openai.ChatCompletions

Chat implements DashScope's chat endpoint.

func NewChat

func NewChat(config ChatConfig) (*Chat, error)

NewChat rejects an invalid provider binding before the first chat call.

type ChatConfig

type ChatConfig struct {
	APIKey         string
	DefaultOptions corechat.Options
	BaseURL        string
	HTTPClient     *http.Client
}

ChatConfig binds provider access and defaults shared by every chat call.

func (ChatConfig) Validate

func (c ChatConfig) Validate() error

type EmbeddingModel

type EmbeddingModel = openai.EmbeddingModel

EmbeddingModel is the shared protocol type itself rather than a wrapper, so this provider adds no second public surface for callers to choose between.

func NewEmbeddingModel

func NewEmbeddingModel(config EmbeddingModelConfig) (*EmbeddingModel, error)

NewEmbeddingModel rejects an invalid provider binding before the first embedding call.

type EmbeddingModelConfig

type EmbeddingModelConfig struct {
	APIKey         string
	DefaultOptions embedding.Options
	BaseURL        string
	HTTPClient     *http.Client
}

EmbeddingModelConfig binds provider access and defaults shared by every embedding call.

func (EmbeddingModelConfig) Validate

func (e EmbeddingModelConfig) Validate() error