GroqCloud provides high-performance AI inference services, enabling developers to deploy and manage AI models efficiently.
GroqCloud provides high-performance AI inference services, enabling developers to deploy and manage AI models efficiently. On Nagent, GroqCloud is exposed as a fully-configurable ai models integration that any agent can call — 7 actions, and API key authentication. No code is required to wire GroqCloud into your workflow — connect it once via the External Integrations panel and reuse it across every agent you build.
Agent builders use GroqCloud to automate the kinds of tasks ai models teams previously handled manually. Concrete examples — each one is a single agent step in Nagent — include:
Every action and trigger is paired with a structured input/output schema (visible in the sections below), so when you wire GroqCloud into Helix — our agentic agent builder — the editor knows exactly what each step expects and produces. Configure once, deploy anywhere across your Nagent agents.
Every operation an agent can call against GroqCloud, with input parameters and output schema. Drop these into any step of an agent built in Helix.
GROQCLOUD_CREATE_AUDIO_TRANSCRIPTIONTool to transcribe audio into text in the same language as the audio. Use when you need to convert speech to text while preserving the original language. Supports multiple formats including mp3, mp4, wav, and webm.
Input parameters
The audio file to transcribe. Supported formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, webm.
Model ID for transcription. whisper-large-v3-turbo is faster, whisper-large-v3 may be more accurate.
Optional text to guide the model's style or continue a previous audio segment. The prompt should match the audio language.
Language of the input audio in ISO-639-1 format (e.g., 'en', 'es', 'fr'). Supplying this will improve accuracy and latency.
Sampling temperature between 0 and 1. Higher values (e.g., 0.8) make output more random, lower values (e.g., 0.2) make it more focused. If set to 0, model uses log probability to auto-adjust temperature.
Output format. Use 'verbose_json' for timestamp information, 'json' for basic text, or 'text' for plain text output.
Timestamp granularities to populate. Requires response_format='verbose_json'. Options: 'word' (adds latency), 'segment' (no additional latency). Can specify both.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_CREATE_RESPONSETool to create a model response for the given input. Beta endpoint with simplified interface compared to chat completions. Use when you need a streamlined API for generating model responses.
Input parameters
Response format configuration.
Optional identifier for tracking end-user requests for monitoring and compliance
Text input to the model or an array of input items
ID of the model to use. See available models at https://console.groq.com/docs/models
Response storage flag. Currently only supports false or null
List of tools available to the model. Maximum of 128 functions
Nucleus sampling parameter controlling cumulative probability cutoff. Range 0-1
Enable streaming mode to receive response data as server-sent events
Custom key-value pairs for storing additional information. Maximum of 16 pairs
Configuration for reasoning capabilities.
Context truncation strategy.
Controls randomness. Range 0-2. Lower is more deterministic, higher is more creative
Controls which tool is called. Values: 'none', 'auto', 'required', or specific function
System message inserted as the first item in the model's context
Service tier for processing the request.
Upper bound for tokens in the response, including visible and reasoning tokens
Enable parallel execution of multiple tool calls
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_GROQ_CREATE_AUDIO_TRANSLATIONTool to translate an audio file into English text. Use when you have a non-English recording and need an accurate English transcript. Use after confirming the file path.
Input parameters
The audio file to translate to English. Supported formats: mp3, wav, etc.
Model ID for translation (e.g., 'whisper-large-v3'). whisper-large-v3-turbo may not support translations.
Optional prompt to guide the translation output.
Sampling temperature between 0.0 and 1.0 to control randomness.
Output format: 'json', 'verbose_json', or 'text'.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_GROQ_CREATE_CHAT_COMPLETIONTool to generate a chat-based completion for a conversation. Use when you have a list of prior messages and need the model's next reply. Response completion text is at choices\[0\].message.content in the returned envelope.
Input parameters
Number of chat completion choices to generate (must be 1)
Up to 4 stop sequences where the model will stop generating further tokens
Unique identifier for the end user for monitoring/abuse detection
ID of the model to use Verify valid IDs via GROQCLOUD_LIST_MODELS before use; hard-coded IDs may be deprecated. Different models have different token limits and rate quotas — check model metadata before large-scale completions.
Nucleus sampling parameter (0 to 1)
Ordered list of messages comprising the conversation
Sampling temperature between 0 and 2
Maximum number of tokens to generate in the completion
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_GROQ_RETRIEVE_MODELTool to retrieve detailed information about a specific model. Use after listing models when you need metadata for a chosen model. Returned metadata may change as models update; do not cache.
Input parameters
Identifier of the model to retrieve Must be an exact ID from GROQCLOUD_LIST_MODELS; approximated or guessed IDs will fail.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_LIST_MODELSTool to list all available models and their metadata. Always call this to retrieve current model IDs rather than using hard-coded or cached identifiers, as deprecated names cause failures in GROQCLOUD_GROQ_RETRIEVE_MODEL and GROQCLOUD_GROQ_CREATE_CHAT_COMPLETION. Returns availability and metadata only — excludes usage stats, latency metrics, and pricing. Response may include many models; filter client-side by provider, family, modality, or context length. Frequent polling combined with high-volume requests risks HTTP 429 rate_limit_exceeded; use backoff and minimize call frequency.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
GROQCLOUD_LIST_VOICESTool to retrieve available TTS voices for Groq PlayAI models. Use when you need to discover voice options before calling text-to-speech. Note: static list maintained manually; no live endpoint exists.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
No publicly available marketplace agent is found using this tool yet. There are 42 agents privately built on Nagent that already use GroqCloud.
Build on Nagent
Connect GroqCloud to any Nagent agent in minutes — no API key management, no boilerplate. Just configure and deploy.
The five questions agent builders ask before adopting a new integration.
Open the External Integrations panel inside Nagent (app.nagent.ai/externalIntegration), find GroqCloud, and click "Connect Now." You'll authenticate with an API key — Nagent handles credential storage and refresh automatically. Once connected, GroqCloud is available to any agent in your workspace.
No. Nagent provides no-code integration for every tool. Once GroqCloud is connected, you configure its 7 actions directly in the agent builder UI — no API calls, no boilerplate, no schema management.
Helix — Nagent's agentic agent builder — lets you drop GroqCloud steps into any workflow visually. Pick an action (e.g., one of those listed above), fill in the inputs (Helix knows the required vs. optional schema for each parameter), and connect it to upstream/downstream steps. Triggers run as the entry point of an agent, so when a GroqCloud event fires, the agent kicks off automatically.
Every GroqCloud action and trigger ships with a fully-typed schema — input parameters with name, type, required flag, and description, plus the output payload shape. The schemas are documented in the sections above. Helix uses these schemas to validate your configuration at build time and to type-check the data flowing between steps.
Yes. While GroqCloud ships with 7 pre-built ai models actions, you can layer custom logic around them inside Helix — pre/post-processing steps, conditional branches, retries, or stitching GroqCloud together with other connected tools. For deeper customization, talk to our team about Nagent's Agentic AI Lab — forward-deployed engineers who build GroqCloud-based workflows tailored to your business.