сделал gpt4

vendor/github.com/sashabaranov/go-openai/stream.go

@@ -0,0 +1,61 @@
+package openai
+
+import (
+	"bufio"
+	"context"
+	"errors"
+	"net/http"
+)
+
+var (
+	ErrTooManyEmptyStreamMessages = errors.New("stream has sent too many empty messages")
+)
+
+type CompletionStream struct {
+	*streamReader[CompletionResponse]
+}
+
+// CreateCompletionStream — API call to create a 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) CreateCompletionStream(
+	ctx context.Context,
+	request CompletionRequest,
+) (stream *CompletionStream, err error) {
+	urlSuffix := "/completions"
+	if !checkEndpointSupportsModel(urlSuffix, request.Model) {
+		err = ErrCompletionUnsupportedModel
+		return
+	}
+
+	if !checkPromptType(request.Prompt) {
+		err = ErrCompletionRequestPromptTypeNotSupported
+		return
+	}
+
+	request.Stream = true
+	req, err := c.newStreamRequest(ctx, "POST", urlSuffix, request)
+	if err != nil {
+		return
+	}
+
+	resp, err := c.config.HTTPClient.Do(req) //nolint:bodyclose // body is closed in stream.Close()
+	if err != nil {
+		return
+	}
+	if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusBadRequest {
+		return nil, c.handleErrorResp(resp)
+	}
+
+	stream = &CompletionStream{
+		streamReader: &streamReader[CompletionResponse]{
+			emptyMessagesLimit: c.config.EmptyMessagesLimit,
+			reader:             bufio.NewReader(resp.Body),
+			response:           resp,
+			errAccumulator:     newErrorAccumulator(),
+			unmarshaler:        &jsonUnmarshaler{},
+		},
+	}
+	return
+}