Groq

Groq builds custom Language Processing Unit (LPU) silicon optimized for low-latency LLM inference. The GroqCloud API serves popular open models (Llama, GPT OSS, Whisper, Orpheus) at industry-leading tokens-per-second with an OpenAI-compatible interface.

13 APIs 0 Features

AILLMInferenceLPULow Latency

APIs

Groq Chat Completions API

OpenAI-compatible chat completions across Llama, GPT OSS, Mixtral, Gemma, and Whisper-family models running on Groq LPU silicon, with streaming, tool use, and structured outputs.

Groq Reasoning API

Reasoning-capable models with explicit chain-of-thought support, surfaced through the chat completions endpoint.

Groq Vision API

Image and document understanding plus OCR via vision-capable chat models.

Groq Speech-to-Text API

OpenAI-compatible audio transcription endpoint serving Whisper-family models on LPU hardware.

Groq Text-to-Speech API

Speech synthesis using Orpheus and other TTS models, billed per million characters.

Groq Content Moderation API

Safety classifier endpoint (Llama Guard) for input/output policy compliance.

Groq Batch API

Asynchronous batch inference at 50% off synchronous rates for non-realtime workloads.

Groq Flex Processing API

Flexible service tier offering higher throughput at relaxed latency targets for cost-sensitive workloads.

Groq Files API

Upload and manage files for batch inputs and other workflows.

Groq Models API

Lists models available on GroqCloud with metadata, context length, and pricing tags.

Groq Tools API

Built-in tools - Web Search, Browser Automation, Code Execution, Wolfram Alpha - invocable from chat completions and billed per call or per hour.

Groq LoRA Inference API

Serves customer LoRA adapters on top of supported base models for low-latency custom inference.

Groq Prompt Caching

Automatic prompt caching with a 50% discount on cached input tokens and no extra caching fee.

Collections

GroqCloud API

OPEN

GraphQL

Groq GraphQL API

Groq provides ultra-fast LLM inference via their Language Processing Unit (LPU) hardware. The API is OpenAI-compatible and covers chat completions, audio transcription, and batc...

GRAPHQL

Pricing Plans

Groq Plans Pricing

4 plans

PLANS

Rate Limits

Groq Rate Limits

6 limits

RATE LIMITS

FinOps

Groq Finops

FINOPS

Event Specifications

Groq Chat Completions Streaming (HTTP + SSE)

AsyncAPI 2.6 description of Groq's **chat completion streaming** surface. Groq does not publish a WebSocket API. The only asynchronous / event-style transport documented at http...

ASYNCAPI

Resources

👥

GitHubOrganization

Sources

opencollection: 1.0.0
info:
  name: GroqCloud API
  version: '2.1'
request:
  auth:
    type: bearer
    token: '{{bearerToken}}'
items:
- info:
    name: Audio
    type: folder
  items:
  - info:
      name: Generates audio from the input text.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/audio/speech
      body:
        type: json
        data: '{}'
    docs: Generates audio from the input text.
  - info:
      name: Transcribes audio into the input language.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/audio/transcriptions
      body:
        type: multipart-form
        data: []
    docs: Transcribes audio into the input language.
  - info:
      name: Translates audio into English.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/audio/translations
      body:
        type: multipart-form
        data: []
    docs: Translates audio into English.
- info:
    name: Batch
    type: folder
  items:
  - info:
      name: List your organization's batches.
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/batches
    docs: List your organization's batches.
  - info:
      name: Creates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch).
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/batches
      body:
        type: json
        data: '{}'
    docs: Creates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch).
  - info:
      name: Retrieves a batch.
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/batches/:batch_id
      params:
      - name: batch_id
        value: ''
        type: path
        description: The ID of the batch to retrieve.
    docs: Retrieves a batch.
  - info:
      name: Cancels a batch.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/batches/:batch_id/cancel
      params:
      - name: batch_id
        value: ''
        type: path
        description: The ID of the batch to cancel.
    docs: Cancels a batch.
- info:
    name: Chat
    type: folder
  items:
  - info:
      name: Creates a model response for the given chat conversation.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/chat/completions
      body:
        type: json
        data: '{}'
    docs: Creates a model response for the given chat conversation.
- info:
    name: Embeddings
    type: folder
  items:
  - info:
      name: Creates an embedding vector representing the input text.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/embeddings
      body:
        type: json
        data: '{}'
    docs: Creates an embedding vector representing the input text.
- info:
    name: Files
    type: folder
  items:
  - info:
      name: Returns a list of files.
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/files
    docs: Returns a list of files.
  - info:
      name: 'Upload a file that can be used across various endpoints.


        The Batch API only supports `.jsonl` files up to 100 MB in size. The input also has a specific required [format](/docs/batch).


        Please contact us if you need to increase these storage limits.

        '
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/files
      body:
        type: multipart-form
        data: []
    docs: 'Upload a file that can be used across various endpoints.


      The Batch API only supports `.jsonl` files up to 100 MB in size. The input also has a specific required [format](/docs/batch).


      Please contact us if you need to increase these storage limits.

      '
  - info:
      name: Returns information about a file.
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/files/:file_id
      params:
      - name: file_id
        value: ''
        type: path
        description: The file to retrieve
    docs: Returns information about a file.
  - info:
      name: Delete a file.
      type: http
    http:
      method: DELETE
      url: https://api.groq.com/openai/v1/files/:file_id
      params:
      - name: file_id
        value: ''
        type: path
        description: The ID of the file to use for this request.
    docs: Delete a file.
  - info:
      name: Returns the contents of the specified file.
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/files/:file_id/content
      params:
      - name: file_id
        value: ''
        type: path
        description: The ID of the file to use for this request.
    docs: Returns the contents of the specified file.
- info:
    name: Fine Tuning
    type: folder
  items:
  - info:
      name: Lists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact)
        for more information.
      type: http
    http:
      method: GET
      url: https://api.groq.com/v1/fine_tunings
    docs: Lists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact)
      for more information.
  - info:
      name: Creates a new fine tuning for the already uploaded files This endpoint is in closed beta. [Contact us](https://groq.com/contact)
        for more information.
      type: http
    http:
      method: POST
      url: https://api.groq.com/v1/fine_tunings
      body:
        type: json
        data: '{}'
    docs: Creates a new fine tuning for the already uploaded files This endpoint is in closed beta. [Contact us](https://groq.com/contact)
      for more information.
  - info:
      name: Retrieves an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact)
        for more information.
      type: http
    http:
      method: GET
      url: https://api.groq.com/v1/fine_tunings/:id
      params:
      - name: id
        value: ''
        type: path
    docs: Retrieves an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact)
      for more information.
  - info:
      name: Deletes an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact)
        for more information.
      type: http
    http:
      method: DELETE
      url: https://api.groq.com/v1/fine_tunings/:id
      params:
      - name: id
        value: ''
        type: path
    docs: Deletes an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for
      more information.
- info:
    name: Models
    type: folder
  items:
  - info:
      name: List all available [models](https://console.groq.com/docs/models).
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/models
    docs: get all available models
  - info:
      name: Get detailed information about a [model](https://console.groq.com/docs/models).
      type: http
    http:
      method: GET
      url: https://api.groq.com/openai/v1/models/:model
      params:
      - name: model
        value: ''
        type: path
        description: The model to get
    docs: Get a specific model
  - info:
      name: Delete model
      type: http
    http:
      method: DELETE
      url: https://api.groq.com/openai/v1/models/:model
      params:
      - name: model
        value: ''
        type: path
        description: The model to delete
    docs: Delete a model
- info:
    name: Reranking
    type: folder
  items:
  - info:
      name: Reranks documents based on their relevance to a query.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/reranking
      body:
        type: json
        data: '{}'
    docs: 'Given a query and a list of documents, returns the documents ranked by their relevance to the query.

      The documents are scored and sorted in descending order of relevance.

      '
- info:
    name: Responses
    type: folder
  items:
  - info:
      name: Creates a model response for the given input.
      type: http
    http:
      method: POST
      url: https://api.groq.com/openai/v1/responses
      body:
        type: json
        data: '{}'
    docs: Creates a model response for the given input.
bundled: true