2054db016c
* Add support for O3-mini - Add support for the o3 mini set of models, including tests that match the constraints in OpenAI's API docs (https://platform.openai.com/docs/models#o3-mini). * Deprecate and refactor - Deprecate `ErrO1BetaLimitationsLogprobs` and `ErrO1BetaLimitationsOther` - Implement `validationRequestForReasoningModels`, which works on both o1 & o3, and has per-model-type restrictions on functionality (eg, o3 class are allowed function calls and system messages, o1 isn't) * Move reasoning validation to `reasoning_validator.go` - Add a `NewReasoningValidator` which exposes a `Validate()` method for a given request - Also adds a test for chat streams * Final nits
107 lines
3.8 KiB
Go
107 lines
3.8 KiB
Go
package openai
|
|
|
|
import (
|
|
"context"
|
|
"net/http"
|
|
)
|
|
|
|
type ChatCompletionStreamChoiceDelta struct {
|
|
Content string `json:"content,omitempty"`
|
|
Role string `json:"role,omitempty"`
|
|
FunctionCall *FunctionCall `json:"function_call,omitempty"`
|
|
ToolCalls []ToolCall `json:"tool_calls,omitempty"`
|
|
Refusal string `json:"refusal,omitempty"`
|
|
}
|
|
|
|
type ChatCompletionStreamChoiceLogprobs struct {
|
|
Content []ChatCompletionTokenLogprob `json:"content,omitempty"`
|
|
Refusal []ChatCompletionTokenLogprob `json:"refusal,omitempty"`
|
|
}
|
|
|
|
type ChatCompletionTokenLogprob struct {
|
|
Token string `json:"token"`
|
|
Bytes []int64 `json:"bytes,omitempty"`
|
|
Logprob float64 `json:"logprob,omitempty"`
|
|
TopLogprobs []ChatCompletionTokenLogprobTopLogprob `json:"top_logprobs"`
|
|
}
|
|
|
|
type ChatCompletionTokenLogprobTopLogprob struct {
|
|
Token string `json:"token"`
|
|
Bytes []int64 `json:"bytes"`
|
|
Logprob float64 `json:"logprob"`
|
|
}
|
|
|
|
type ChatCompletionStreamChoice struct {
|
|
Index int `json:"index"`
|
|
Delta ChatCompletionStreamChoiceDelta `json:"delta"`
|
|
Logprobs *ChatCompletionStreamChoiceLogprobs `json:"logprobs,omitempty"`
|
|
FinishReason FinishReason `json:"finish_reason"`
|
|
ContentFilterResults ContentFilterResults `json:"content_filter_results,omitempty"`
|
|
}
|
|
|
|
type PromptFilterResult struct {
|
|
Index int `json:"index"`
|
|
ContentFilterResults ContentFilterResults `json:"content_filter_results,omitempty"`
|
|
}
|
|
|
|
type ChatCompletionStreamResponse struct {
|
|
ID string `json:"id"`
|
|
Object string `json:"object"`
|
|
Created int64 `json:"created"`
|
|
Model string `json:"model"`
|
|
Choices []ChatCompletionStreamChoice `json:"choices"`
|
|
SystemFingerprint string `json:"system_fingerprint"`
|
|
PromptAnnotations []PromptAnnotation `json:"prompt_annotations,omitempty"`
|
|
PromptFilterResults []PromptFilterResult `json:"prompt_filter_results,omitempty"`
|
|
// An optional field that will only be present when you set stream_options: {"include_usage": true} in your request.
|
|
// When present, it contains a null value except for the last chunk which contains the token usage statistics
|
|
// for the entire request.
|
|
Usage *Usage `json:"usage,omitempty"`
|
|
}
|
|
|
|
// ChatCompletionStream
|
|
// Note: Perhaps it is more elegant to abstract Stream using generics.
|
|
type ChatCompletionStream struct {
|
|
*streamReader[ChatCompletionStreamResponse]
|
|
}
|
|
|
|
// CreateChatCompletionStream — API call to create a chat completion w/ streaming
|
|
// support. It sets whether to stream back partial progress. If set, tokens will be
|
|
// sent as data-only server-sent events as they become available, with the
|
|
// stream terminated by a data: [DONE] message.
|
|
func (c *Client) CreateChatCompletionStream(
|
|
ctx context.Context,
|
|
request ChatCompletionRequest,
|
|
) (stream *ChatCompletionStream, err error) {
|
|
urlSuffix := chatCompletionsSuffix
|
|
if !checkEndpointSupportsModel(urlSuffix, request.Model) {
|
|
err = ErrChatCompletionInvalidModel
|
|
return
|
|
}
|
|
|
|
request.Stream = true
|
|
reasoningValidator := NewReasoningValidator()
|
|
if err = reasoningValidator.Validate(request); err != nil {
|
|
return
|
|
}
|
|
|
|
req, err := c.newRequest(
|
|
ctx,
|
|
http.MethodPost,
|
|
c.fullURL(urlSuffix, withModel(request.Model)),
|
|
withBody(request),
|
|
)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
resp, err := sendRequestStream[ChatCompletionStreamResponse](c, req)
|
|
if err != nil {
|
|
return
|
|
}
|
|
stream = &ChatCompletionStream{
|
|
streamReader: resp,
|
|
}
|
|
return
|
|
}
|