> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zerogpu.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# deepseek-v4-flash

> Model details for deepseek-v4-flash. Reasoning, coding, and agentic workflows with a 1M-token context window.

<Note>
  This model supports the Chat Completions API only. Send requests to
  `/v1/chat/completions` — the Responses endpoint (`/v1/responses`) is not
  available for this model.
</Note>

> DeepSeek's DeepSeek-V4-Flash is an open-weight Mixture-of-Experts model built for efficient reasoning, coding, and agentic workflows, with 284B total parameters activating only 13B per token, served on ZeroGPU for general text generation. It sustains a 1,048,576-token (1M) context well suited for analyzing large codebases, long documents, extensive conversations, and complex research tasks, and supports fast non-thinking responses as well as higher-effort reasoning modes for planning, problem-solving, and multi-step automation. MIT-licensed with no usage restrictions.

**References:** [Model docs](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash) • [Terms](https://zerogpu.ai/terms) • [Privacy](https://zerogpu.ai/privacy-policy)


## OpenAPI

````yaml api-reference/openapi/playgrounds/deepseek-v4-flash.openapi.json POST /chat/completions
openapi: 3.1.0
info:
  title: deepseek-v4-flash playground
  version: '1.0'
  description: >-
    Interactive playground for **deepseek-v4-flash**.

    Model is always `deepseek-v4-flash` on this page (shown in the form, not
    editable).

    Use the **request example** selector to switch use cases (JSON, NER, PII,
    etc.).

    Authentication: `x-api-key` (required) and `x-project-id` (optional).
servers:
  - url: https://api.zerogpu.ai/v1
    description: Production
security:
  - ApiKey: []
paths:
  /chat/completions:
    post:
      tags:
        - deepseek-v4-flash
      summary: 'deepseek-v4-flash: Chat completions'
      operationId: createChatCompletion_deepseek-v4-flash
      parameters:
        - name: x-project-id
          in: header
          required: false
          schema:
            type: string
          description: >-
            Optional project identifier. Scopes the request to a specific
            project when provided.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              allOf:
                - $ref: '#/components/schemas/CreateChatCompletionRequest'
                - type: object
                  properties:
                    model:
                      type: string
                      const: deepseek-v4-flash
                      default: deepseek-v4-flash
                      example: deepseek-v4-flash
                      description: >-
                        Model identifier (fixed for this playground). Use
                        request examples to change use cases.
            examples:
              default:
                summary: Default
                value:
                  model: deepseek-v4-flash
                  messages:
                    - role: system
                      content: You are a pragmatic staff engineer. Be concise.
                    - role: user
                      content: >-
                        Plan the migration of a nightly cron-based ETL job to an
                        event-driven pipeline. 4 steps max.
                  max_tokens: 800
      responses:
        '200':
          description: Success
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ChatCompletionResponse'
              examples:
                default:
                  summary: Default
                  value:
                    id: id-1785523944762
                    object: chat.completion
                    created: 1785523944
                    model: deepseek-v4-flash
                    choices:
                      - index: 0
                        finish_reason: stop
                        message:
                          role: assistant
                          content: >-
                            **1. Inventory the job** – Document inputs, outputs,
                            schedule, and every downstream consumer of the cron
                            ETL so nothing breaks silently.


                            **2. Define the events** – Pick the real trigger
                            (file landing, DB change, message) and publish it to
                            a queue or event bus with a versioned schema.


                            **3. Run in parallel** – Build the event-driven
                            consumer alongside cron, write to a shadow table,
                            and diff outputs until they match.


                            **4. Cut over with a rollback path** – Point
                            consumers at the new output, keep cron paused (not
                            deleted) for one cycle, then decommission.
                          reasoning: >-
                            Cron ETL to event-driven: inventory the current job
                            first, define the trigger events, run the new
                            pipeline in parallel against a shadow output, then
                            cut over with a rollback path. Four steps, each
                            actionable.
                          tool_calls: []
                    usage:
                      prompt_tokens: 52
                      completion_tokens: 158
                      total_tokens: 210
        '400':
          description: Bad request
        '401':
          description: Unauthorized
        '403':
          description: Forbidden
        '420':
          description: Insufficient quota
        '500':
          description: Internal server error
components:
  schemas:
    CreateChatCompletionRequest:
      type: object
      required:
        - model
        - messages
      properties:
        model:
          type: string
          const: deepseek-v4-flash
          default: deepseek-v4-flash
          example: deepseek-v4-flash
          description: >-
            Model identifier (fixed for this playground). Use request examples
            to change use cases.
        messages:
          type: array
          minItems: 1
          items:
            $ref: '#/components/schemas/ChatMessage'
        max_tokens:
          type: integer
          minimum: 1
          example: 800
          description: Maximum number of tokens to generate in the response.
        metadata:
          type: object
          additionalProperties: true
    ChatCompletionResponse:
      type: object
      additionalProperties: true
    ChatMessage:
      type: object
      required:
        - role
        - content
      properties:
        role:
          type: string
          enum:
            - system
            - user
            - assistant
        content:
          type: string
          minLength: 1
          format: textarea
          maxLength: 131072
          description: Message text.
  securitySchemes:
    ApiKey:
      type: apiKey
      in: header
      name: x-api-key

````