Update the URL and API desc (#785)

janhq · Jun 28, 2024 · bd8968a · bd8968a
1 parent cc5d7ce
commit bd8968a
Show file tree

Hide file tree

Showing 7 changed files with 26 additions and 9 deletions.
diff --git a/cortex-js/src/app.ts b/cortex-js/src/app.ts
@@ -63,10 +63,17 @@ export const getApp = async () => {
     )
     .addTag(
       'Events',
-      'Endpoints for  observing Cortex statuses through event notifications.',
+      'Endpoints for observing Cortex statuses through event notifications.',
+    )
+    .addTag(
+      'Configurations',
+      "Endpoints for customizing the Cortex's configurations.",
+    )
+    .addTag(
+      'Engines',
+      'Endpoints for managing the available engines within Cortex.',
     )
     .addServer('http://localhost:1337')
-    .addServer('http://localhost:1337/v1')
     .build();
   const document = SwaggerModule.createDocument(app, config);
 

diff --git a/cortex-js/src/infrastructure/controllers/chat.controller.ts b/cortex-js/src/infrastructure/controllers/chat.controller.ts
@@ -12,7 +12,7 @@ export class ChatController {
 
   @ApiOperation({
     summary: 'Create chat completion',
-    description: 'Creates a model response for the given conversation.',
+    description: 'Creates a model response for the given conversation. The following parameters are not working for the `TensorRT-LLM` engine:\n- `frequency_penalty`\n- `presence_penalty`\n- `top_p`',
   })
   @HttpCode(200)
   @ApiResponse({

diff --git a/cortex-js/src/infrastructure/controllers/models.controller.ts b/cortex-js/src/infrastructure/controllers/models.controller.ts
@@ -115,6 +115,8 @@ export class ModelsController {
   }
 
   @ApiOperation({
+    summary: 'Abort model download',
+    description: 'Abort the model download operation.',
     parameters: [
       {
         in: 'path',
@@ -136,7 +138,7 @@ export class ModelsController {
     type: DownloadModelResponseDto,
   })
   @ApiOperation({
-    summary: 'Pulls a remote model and download it',
+    summary: 'Download a remote model',
     description:
       'Pulls a remote model template from cortex hub or huggingface and downloads it.',
   })

diff --git a/cortex-js/src/infrastructure/dtos/chat/chat-completion-message.dto.ts b/cortex-js/src/infrastructure/dtos/chat/chat-completion-message.dto.ts
@@ -2,7 +2,7 @@ import { IsString } from 'class-validator';
 import { ApiProperty } from '@nestjs/swagger';
 
 export class ChatCompletionMessage {
-  @ApiProperty({ description: 'The Content of the chat message.' })
+  @ApiProperty({ description: 'The Content of the chat message.',  })
   @IsString()
   content: string;
 

diff --git a/cortex-js/src/infrastructure/dtos/chat/create-chat-completion.dto.ts b/cortex-js/src/infrastructure/dtos/chat/create-chat-completion.dto.ts
@@ -22,13 +22,15 @@ export class CreateChatCompletionDto {
 
   @ApiProperty({
     description: 'The unique identifier of the model.',
+    example: 'gpt-4',
   })
   @IsString()
   model: string;
 
   @ApiProperty({
     description:
       'Determines the format for output generation. If set to `true`, the output is generated continuously, allowing for real-time streaming of responses. If set to `false`, the output is delivered in a single JSON file.',
+      example: true,
   })
   @IsOptional()
   @IsBoolean()
@@ -37,6 +39,7 @@ export class CreateChatCompletionDto {
   @ApiProperty({
     description:
       'Sets the upper limit on the number of tokens the model can generate in a single output.',
+      example: 4096,
   })
   @IsOptional()
   @IsNumber()
@@ -45,6 +48,7 @@ export class CreateChatCompletionDto {
   @ApiProperty({
     description:
       'Defines specific tokens or phrases that signal the model to stop producing further output.',
+      example: ["End"],
   })
   @IsOptional()
   @IsArray()
@@ -53,6 +57,7 @@ export class CreateChatCompletionDto {
   @ApiProperty({
     description:
       'Modifies the likelihood of the model repeating the same words or phrases within a single output.',
+      example: 0.2,
   })
   @IsOptional()
   @IsNumber()
@@ -61,20 +66,23 @@ export class CreateChatCompletionDto {
   @ApiProperty({
     description:
       'Reduces the likelihood of repeating tokens, promoting novelty in the output.',
+      example: 0.6,
   })
   @IsOptional()
   @IsNumber()
   presence_penalty?: number;
 
   @ApiProperty({
     description: "Influences the randomness of the model's output.",
+    example: 0.8,
   })
   @IsOptional()
   @IsNumber()
   temperature?: number;
 
   @ApiProperty({
     description: 'Sets probability threshold for more relevant outputs.',
+    example: 0.95,
   })
   @IsOptional()
   @IsNumber()

diff --git a/cortex-js/src/infrastructure/dtos/common/common-response.dto.ts b/cortex-js/src/infrastructure/dtos/common/common-response.dto.ts
@@ -3,7 +3,7 @@ import { IsString } from 'class-validator';
 
 export class CommonResponseDto {
   @ApiProperty({
-    description: 'The success or error message',
+    description: 'The response success or error message.',
   })
   @IsString()
   message: string;

diff --git a/cortex-js/src/infrastructure/dtos/engines/engines.dto.ts b/cortex-js/src/infrastructure/dtos/engines/engines.dto.ts
@@ -16,7 +16,7 @@ export class EngineDto implements Partial<Extension> {
   @ApiProperty({
     type: String,
     example: 'Cortex',
-    description: 'The display name of the engine',
+    description: 'The display name of the engine.',
   })
   @IsString()
   @IsOptional()
@@ -25,7 +25,7 @@ export class EngineDto implements Partial<Extension> {
   @ApiProperty({
     type: String,
     example: 'Cortex engine',
-    description: 'The description of the engine',
+    description: 'The description of the engine.',
   })
   @IsString()
   @IsOptional()
@@ -34,7 +34,7 @@ export class EngineDto implements Partial<Extension> {
   @ApiProperty({
     type: String,
     example: '0.0.1',
-    description: 'The version of the engine',
+    description: 'The version of the engine.',
   })
   @IsString()
   @IsOptional()