Add GuidedChoice to ChatCompletionRequest (sashabaranov#1034)

* Add GuidedChoice to ChatCompletionRequest

* made separate NonOpenAIExtensions

* fixed lint issue

* renamed struct and removed inline json tag

* Update chat.go

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* Update chat.go

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Alex Baranov <677093+sashabaranov@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

Author: 3 people

chat.go

@@ -279,6 +279,16 @@ func (r *ChatCompletionResponseFormatJSONSchema) UnmarshalJSON(data []byte) erro
 	return nil
 }
 
+// ChatCompletionRequestExtensions contains third-party OpenAI API extensions (e.g., vendor-specific implementations like vLLM).
+type ChatCompletionRequestExtensions struct {
+	// GuidedChoice is a vLLM-specific extension that restricts the model's output
+	// to one of the predefined string choices provided in this field. This feature
+	// is used to constrain the model's responses to a controlled set of options,
+	// ensuring predictable and consistent outputs in scenarios where specific
+	// choices are required.
+	GuidedChoice []string `json:"guided_choice,omitempty"`
+}
+
 // ChatCompletionRequest represents a request structure for chat completion API.
 type ChatCompletionRequest struct {
 	Model    string                  `json:"model"`
@@ -340,6 +350,8 @@ type ChatCompletionRequest struct {
 	ChatTemplateKwargs map[string]any `json:"chat_template_kwargs,omitempty"`
 	// Specifies the latency tier to use for processing the request.
 	ServiceTier ServiceTier `json:"service_tier,omitempty"`
+	// Embedded struct for non-OpenAI extensions
+	ChatCompletionRequestExtensions
 }
 
 type StreamOptions struct {