The highest tagged major version is
Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
var ( // ErrResultTruncated is returned when the OpenAI API returned a truncated // result. The reason for the truncation will be appended to the error // string. ErrResultTruncated = errors.New("result was truncated") // ErrNoResults is returned if the OpenAI API returned an empty result. This // should not generally happen. ErrNoResults = errors.New("no results return from API") // ErrUnsupportedBackend is returned if the provided backend name is // unknown. ErrUnsupportedBackend = errors.New("unsupported backend") // ErrUnsupportedModel is returned if the SetModel method is provided with // an unsupported model ErrUnsupportedModel = errors.New("unsupported model") // ErrUnexpectedStatus is returned when the OpenAI API returned a response // with an unexpected status code ErrUnexpectedStatus = errors.New("OpenAI returned unexpected response") // ErrRequestFailed is returned when the OpenAI API returned an error for // the request ErrRequestFailed = errors.New("request failed") )
Functions ¶
func ExtractCode ¶
ExtractCode receives the full output string from the OpenAI API and attempts to extract a code block from it. OpenAI code blocks are generally Markdown blocks surrounded by the ``` string on both sides. If successful, the code string will be returned together with a true value, otherwise an empty string is returned together with a false value.
Types ¶
type Backend ¶
type Backend interface {
// ListModels returns a list of all models supported by the backend.
ListModels() []Model
// DefaultModel returns the default model that should be used in the abscence
// of a specific choice by the user.
DefaultModel() Model
// Complete send a prompt to a completion model.
Complete(context.Context, Model, string) (Response, error)
// Chat initiates a conversation with a chat model.
Chat(Model) Conversation
}
Backend is an interface that must be implemented in order to support an LLM provider.
type Conversation ¶
type Conversation interface {
// Send sends a message to the model and returns the response.
Send(context.Context, string, ...Message) (Response, error)
}
Conversation is an interface that must be implemented in order to support chat models in an LLM provider.
type Message ¶
type Message struct {
// Role is the type of the participant. The user is named "user" (in Amazon
// Bedrock, this is equivalent to the "Human" identifier). Anything else is
// considered the AI model.
Role string `json:"role"`
// Content is the text content of the message.
Content string `json:"content"`
}
Message represents a single message in an exchange between a user and an AI model, either as part of a chat or a single completion request.
type Response ¶
type Response struct {
// FullOutput is the complete output returned by the API. This is generally
// a Markdown-formatted Message that contains the generated code, plus
// explanations, if any.
FullOutput string
// Code is the extracted code section from the complete output. If code was
// not found or extraction otherwise failed, this will be the same as
// FullOutput.
Code string
// APIKeyUsed is the API key used when making the request.
APIKeyUsed string
// TokensUsed is the number of tokens utilized by the request. This is
// the "usage.total_tokens" value returned from the API.
TokensUsed int64
}
Response is the struct returned from methods generating code via the OpenAI API.
Source Files