Connect to Groq for ultra-low-latency chat completions on open models (Llama, Mixtral), model discovery, Whisper audio transcription and translation, PlayAI text-to-speech, file management, and the asynchronous Batch API.
| Action | Description | Parameters | Returns |
|---|---|---|---|
cancel_batch ⚠ | Cancel a batch job | batch_id | Batch |
chat_completion | Create a chat completion | model, messages [temperature, max_tokens, top_p] | ChatCompletion |
create_batch | Create a batch job | input_file_id, endpoint [completion_window, metadata] | Batch |
create_speech | Generate speech audio from text | model, input, voice [response_format, sample_rate, speed] | bytes |
delete_file ⚠ | Delete an uploaded file | file_id | bool |
get_batch | Get a batch job by ID | batch_id | Batch |
get_file | Get file metadata by ID | file_id | GroqFile |
get_file_content | Get file content by ID | file_id | bytes |
get_model | Get a model by ID | model | GroqModel |
list_batches | List batch jobs | [limit] | list |
list_files | List uploaded files | [purpose] | list |
list_models | List available models | — | list |
transcribe_audio | Transcribe audio to text | model, file_url_or_bytes [language, response_format] | AudioTranscription |
translate_audio | Translate audio to English text | model, file_url_or_bytes [response_format] | AudioTranscription |
upload_file | Upload a file to Groq | file_content [purpose, filename] | GroqFile |
Full schema explorer, code generation, and live API testing