vllmclient

Documentation

Index

• type Client
  • func (c *Client) LoadLoraAdapter(ctx context.Context, addr string, req LoadAdapterRequest) error
  • func (c *Client) UnloadLoraAdapter(ctx context.Context, addr string, req UnloadAdapterRequest) error
• type LoadAdapterRequest
• type LoadAdapterRequestOptions
• type UnloadAdapterRequest
• type UnloadAdapterRequestOptions

Types

type Client

type Client struct {
	HTTPClient *http.Client
}

func (*Client) LoadLoraAdapter

func (c *Client) LoadLoraAdapter(ctx context.Context, addr string, req LoadAdapterRequest) error

Load a LoRa adapter into the VLLM model server. See: https://docs.vllm.ai/en/latest/models/lora.html#dynamically-serving-lora-adapters

func (*Client) UnloadLoraAdapter

func (c *Client) UnloadLoraAdapter(ctx context.Context, addr string, req UnloadAdapterRequest) error

Unload a LoRa adapter from the VLLM model server. See: https://docs.vllm.ai/en/latest/models/lora.html#dynamically-serving-lora-adapters

type LoadAdapterRequest

type LoadAdapterRequest struct {
	LoraName string `json:"lora_name"`
	LoraPath string `json:"lora_path"`

	Options LoadAdapterRequestOptions `json:"-"`
}

type LoadAdapterRequestOptions

type LoadAdapterRequestOptions struct {
	IgnoreAlreadyLoaded bool
}

type UnloadAdapterRequest

type UnloadAdapterRequest struct {
	LoraName string `json:"lora_name"`

	Options UnloadAdapterRequestOptions `json:"-"`
}

type UnloadAdapterRequestOptions

type UnloadAdapterRequestOptions struct {
	IgnoreNotFound bool
}