support gpt-4-vision-preview model

[sotowang]

Please provide support for the gpt-4-vision-preview model. The ChatCompletionMessage is currently not useful for gpt-4-vision-preview.

[xiaoyi510]

+1

[rkintzi]

Duplicates #539 Awaiting PR: #580

[WhiteNightMo]

+1

[NicksonYap]

yeapp, looking forward for vision support as well!

[anhao]

When will the new version be released?

[NicksonYap]

@anhao it seems to have been merged https://github.com/sashabaranov/go-openai/pull/580

but no docs yet

[yashrao99]

Not sure if people have updated the docs, but here is a simple working example uploading a base64 image to the vision API!

		Model:     openai.GPT4VisionPreview,
		MaxTokens: 1000,
		Messages: []openai.ChatCompletionMessage{
			{
				Role: openai.ChatMessageRoleUser,
				MultiContent: []openai.ChatMessagePart{
					{
						Type: openai.ChatMessagePartTypeText,
						Text: "What's in this image?",
					},
					{
						Type: openai.ChatMessagePartTypeImageURL,
						ImageURL: &openai.ChatMessageImageURL{
							URL:    fmt.Sprintf("data:image/jpg;base64,%s", imgString),
							Detail: openai.ImageURLDetailLow,
						},
					},
				},
			},
		},
	})```