Merge pull request #100 from togethercomputer/batchAPI

add batch API

Author: Anirudh1097

openapi.yaml

@@ -1166,6 +1166,130 @@ paths:
               schema:
                 $ref: '#/components/schemas/SessionListResponse'
           description: List Response
+  /batches:
+    get:
+      tags: ['batches']
+      summary: List batch jobs
+      description: List all batch jobs for the authenticated user
+      security:
+        - bearerAuth: []
+      responses:
+        '200':
+          description: OK
+          content:
+            application/json:
+              schema:
+                type: array
+                items:
+                  $ref: '#/components/schemas/BatchJob'
+        '401':
+          description: Unauthorized
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '500':
+          description: Internal Server Error
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+    post:
+      tags: ['batches']
+      summary: Create a batch job
+      description: Create a new batch job with the given input file and endpoint
+      security:
+        - bearerAuth: []
+      requestBody:
+        required: true
+        content:
+          application/json:
+            schema:
+              $ref: '#/components/schemas/CreateBatchRequest'
+      responses:
+        '201':
+          description: Job created (potentially with warnings)
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchJobWithWarning'
+        '400':
+          description: Bad Request
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '401':
+          description: Unauthorized
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '429':
+          description: Too Many Requests
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '500':
+          description: Internal Server Error
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+
+  /batches/{id}:
+    get:
+      tags: ['batches']
+      summary: Get a batch job
+      description: Get details of a batch job by ID
+      security:
+        - bearerAuth: []
+      parameters:
+        - name: id
+          in: path
+          required: true
+          description: Job ID
+          schema:
+            type: string
+          example: "batch_job_abc123def456"
+      responses:
+        '200':
+          description: OK
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchJob'
+        '400':
+          description: Bad Request
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '401':
+          description: Unauthorized
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '403':
+          description: Forbidden
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '404':
+          description: Not Found
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
+        '500':
+          description: Internal Server Error
+          content:
+            application/json:
+              schema:
+                $ref: '#/components/schemas/BatchErrorResponse'
 
 components:
   securitySchemes:
@@ -3799,3 +3923,103 @@ components:
         - type
         - data
       title: StreamOutput
+
+    CreateBatchRequest:
+      type: object
+      required: [endpoint, input_file_id]
+      properties:
+        endpoint:
+          type: string
+          description: The endpoint to use for batch processing
+          example: "/v1/chat/completions"
+        input_file_id:
+          type: string
+          description: ID of the uploaded input file containing batch requests
+          example: "file-abc123def456ghi789"
+        completion_window:
+          type: string
+          description: Time window for batch completion (optional)
+          example: "24h"
+        priority:
+          type: integer
+          description: Priority for batch processing (optional)
+          example: 1
+        model_id:
+          type: string
+          description: "Model to use for processing batch requests"
+          example: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo"
+    BatchErrorResponse:
+      type: object
+      properties:
+        error:
+          type: string
+    BatchJobWithWarning:
+      type: object
+      properties:
+        job:
+          $ref: '#/components/schemas/BatchJob'
+        warning:
+          type: string
+    BatchJob:
+      type: object
+      properties:
+        id:
+          type: string
+          format: uuid
+          example: "01234567-8901-2345-6789-012345678901"
+        user_id:
+          type: string
+          example: "user_789xyz012"
+        input_file_id:
+          type: string
+          example: "file-input123abc456def"
+        file_size_bytes:
+          type: integer
+          format: int64
+          example: 1048576
+          description: "Size of input file in bytes"
+        status:
+          $ref: '#/components/schemas/BatchJobStatus'
+        job_deadline:
+          type: string
+          format: date-time
+          example: "2024-01-15T15:30:00Z"
+        created_at:
+          type: string
+          format: date-time
+          example: "2024-01-15T14:30:00Z"
+        endpoint:
+          type: string
+          example: "/v1/chat/completions"
+        progress:
+          type: number
+          format: float64
+          example: 75.0
+          description: "Completion progress (0.0 to 100)"
+        model_id:
+          type: string
+          example: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo"
+          description: "Model used for processing requests"
+        output_file_id:
+          type: string
+          example: "file-output789xyz012ghi"
+        error_file_id:
+          type: string
+          example: "file-errors456def789jkl"
+        error:
+          type: string
+        completed_at:
+          type: string
+          format: date-time
+          example: "2024-01-15T15:45:30Z"
+    BatchJobStatus:
+      type: string
+      enum:
+        - VALIDATING
+        - IN_PROGRESS
+        - COMPLETED
+        - FAILED
+        - EXPIRED
+        - CANCELLED
+      example: "IN_PROGRESS"
+      description: "Current status of the batch job"