Skip to content

OmniLLM

Qwen (Alibaba Cloud)

plexusone/omnillm

Qwen (Alibaba Cloud)¶

Overview¶

Models: Qwen3 Max, QwQ, Qwen3.5, Qwen3, Qwen2.5 series
Features: Chat completions, streaming, thinking modes, wide global availability

Configuration¶

client, err := omnillm.NewClient(omnillm.ClientConfig{
    Providers: []omnillm.ProviderConfig{
        {Provider: omnillm.ProviderNameQwen, APIKey: "your-qwen-api-key"},
    },
})

Available Models¶

Flagship Models¶

Model	Description
`qwen3-max`	Latest flagship with thinking capability
`qwen3-max-preview`	Preview with extended thinking
`qwen-max`	Flagship reasoning (International/China)

Balanced Models¶

Model	Description
`qwen3.5-plus`	Advanced balanced with thinking mode
`qwen-plus`	Balanced performance (wide availability)
`qwen3.5-flash`	Ultra-fast lightweight
`qwen-flash`	Fast with context caching

Deep Reasoning (QwQ)¶

Model	Description
`qwq-plus`	Deep reasoning with extended chain-of-thought
`qwq-32b`	Open-source 32B reasoning model

Open Source Models¶

Model	Parameters	Description
`qwen3.5-397b-a17b`	397B	Largest open-source model
`qwen3-235b-a22b`	235B	Dual-mode thinking/non-thinking
`qwen3-32b`	32B	Versatile dual-mode model
`qwen2.5-72b-instruct`	72B	Large instruction-following
`qwen2.5-14b-instruct-1m`	14B	Extended 1M token context

OpenAI Compatibility¶

Qwen uses an OpenAI-compatible API via DashScope:

response, err := client.CreateChatCompletion(ctx, &omnillm.ChatCompletionRequest{
    Model: "qwen-plus",
    Messages: []omnillm.Message{
        {Role: omnillm.RoleUser, Content: "Hello!"},
    },
    Temperature: &[]float64{0.7}[0],
    MaxTokens:   &[]int{1000}[0],
})

Streaming¶

stream, err := client.CreateChatCompletionStream(ctx, &omnillm.ChatCompletionRequest{
    Model: "qwen-plus",
    Messages: messages,
})
if err != nil {
    log.Fatal(err)
}
defer stream.Close()

for {
    chunk, err := stream.Recv()
    if err == io.EOF {
        break
    }
    if err != nil {
        log.Fatal(err)
    }
    fmt.Print(chunk.Choices[0].Delta.Content)
}

Regional Availability¶

Qwen models have wide availability including US regions:

Model	Regions
`qwen-plus`	International, US, China
`qwen-flash`	International, US, China
`qwen-max`	International, China

API Documentation¶