Merge pull request #6 from launchableinc/support-json-mode

Support json mode

Author: Konboi

api/src/main/java/com/launchableinc/openai/completion/chat/ChatCompletionRequest.java

@@ -27,6 +27,14 @@ public class ChatCompletionRequest {
 	 */
 	List<ChatMessage> messages;
 
+	/**
+	 * An object specifying the format that the model must output. Compatible with GPT-4 Turbo and all
+	 * GPT-3.5 Turbo models newer than gpt-3.5-turbo-1106.
+	 * https://platform.openai.com/docs/api-reference/chat/create#chat-create-response_format
+	 */
+	@JsonProperty("response_format")
+	ChatResponseFormat responseFormat;
+
 	/**
 	 * What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output
 	 * more random, while lower values like 0.2 will make it more focused and deterministic.<br> We

api/src/main/java/com/launchableinc/openai/completion/chat/ChatResponseFormat.java

@@ -0,0 +1,31 @@
+package com.launchableinc.openai.completion.chat;
+
+import com.fasterxml.jackson.annotation.JsonValue;
+import lombok.Builder;
+import lombok.Data;
+
+/*
+ * OpenAI API Document:https://platform.openai.com/docs/api-reference/chat/create#chat-create-response_format
+ */
+@Data
+@Builder
+public class ChatResponseFormat {
+
+	private ResponseFormat type;
+
+	public enum ResponseFormat {
+		TEXT("text"), JSON("json_object");
+
+		private final String value;
+
+		ResponseFormat(String value) {
+			this.value = value;
+		}
+
+		@JsonValue
+		public String getValue() {
+			return value;
+		}
+	}
+
+}

service/src/test/java/com/launchableinc/openai/service/ChatCompletionTest.java

@@ -2,9 +2,12 @@
 
 import com.fasterxml.jackson.annotation.JsonProperty;
 import com.fasterxml.jackson.annotation.JsonPropertyDescription;
+import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.databind.ObjectMapper;
 import com.fasterxml.jackson.databind.node.ObjectNode;
 import com.launchableinc.openai.completion.chat.*;
+import com.launchableinc.openai.completion.chat.ChatResponseFormat.ResponseFormat;
 import org.junit.jupiter.api.Assumptions;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Test;
@@ -77,6 +80,37 @@ void createChatCompletion() {
 		assertEquals(5, choices.size());
 	}
 
+	@Test
+	void createChatCompletion_with_json_mode() {
+		final List<ChatMessage> messages = new ArrayList<>();
+		final ChatMessage systemMessage = new ChatMessage(ChatMessageRole.SYSTEM.value(),
+				"Generate a random name and age json object. name field is a object that has first and last fields. age is a number.");
+		messages.add(systemMessage);
+
+		ChatCompletionRequest chatCompletionRequest = ChatCompletionRequest
+				.builder()
+				.model("gpt-3.5-turbo-1106")
+				.messages(messages)
+				.maxTokens(50)
+				.logitBias(new HashMap<>())
+				.responseFormat(ChatResponseFormat.builder().type(ResponseFormat.JSON).build())
+				.build();
+
+		ChatCompletionChoice choices = service.createChatCompletion(chatCompletionRequest)
+				.getChoices().get(0);
+		assertTrue(isValidJson(choices.getMessage().getContent()));
+	}
+
+	private boolean isValidJson(String jsonString) {
+		ObjectMapper objectMapper = new ObjectMapper();
+		try {
+			objectMapper.readTree(jsonString);
+			return true;
+		} catch (JsonProcessingException e) {
+			return false;
+		}
+	}
+
 	@Test
 	void streamChatCompletion() {
 		final List<ChatMessage> messages = new ArrayList<>();