OpenRouter is a platform that provides a unified API for accessing various large language models (LLMs) from different providers, allowing developers to integrate multiple AI models seamlessly.
OpenRouter is a platform that provides a unified API for accessing various large language models (LLMs) from different providers, allowing developers to integrate multiple AI models seamlessly. On Nagent, Openrouter is exposed as a fully-configurable artificial intelligence integration that any agent can call — 13 actions, and API key authentication. No code is required to wire Openrouter into your workflow — connect it once via the External Integrations panel and reuse it across every agent you build.
Agent builders use Openrouter to automate the kinds of tasks artificial intelligence 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 Openrouter 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 Openrouter, with input parameters and output schema. Drop these into any step of an agent built in Helix.
OPENROUTER_CREATE_CHAT_COMPLETIONTool to generate a chat-style completion. Use after assembling messages and selecting a model. Supports streaming and function calls. Response format varies across models; use explicit prompt instructions to standardize output. Provider-level rate limits and moderation policies differ per model.
Input parameters
Number of completion choices to generate.
Sequences where the API will stop generating further tokens.
A unique identifier for the end-user to help detect abuse.
ID of the model to use for the chat completion. Must exactly match an ID returned by OPENROUTER_LIST_AVAILABLE_MODELS; typos or approximate names cause request failures.
Nucleus sampling probability, between 0 and 1.
Whether to stream back partial message deltas.
List of conversation messages to provide as context.
List of functions the model may call. Only supported by models that explicitly advertise function-calling capability via OPENROUTER_LIST_AVAILABLE_MODELS. Each function schema must be valid JSON; malformed schemas cause runtime failures regardless of prompt correctness.
Adjust the likelihood of specified tokens.
Maximum number of tokens to generate.
Sampling temperature to use, between 0 and 2.
Control function call behavior: 'none', 'auto', or specify {'name': function_name}.
Penalty for new tokens based on presence in the text so far.
Penalty for new tokens based on frequency in the text so far.
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
OPENROUTER_CREATE_COINBASE_CHARGETool to create a Coinbase charge for crypto payment to add credits to your OpenRouter account. Use when you need to purchase credits using cryptocurrency. Returns calldata needed to fulfill the transaction on the specified blockchain.
Input parameters
Amount in USD to charge. This will be converted to the equivalent crypto amount.
Ethereum wallet address of the sender (must be a valid 42-character hex address starting with '0x').
The app display name that customizes how your app appears in OpenRouter's dashboard.
Blockchain chain ID to use for the transaction. Options: 1 (Ethereum), 137 (Polygon), 8453 (Base).
The app identifier (your app's URL) used as the primary identifier for rankings and to track API usage per application.
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
OPENROUTER_CREATE_MESSAGETool to create a message using Anthropic Messages API format via OpenRouter. Use when you need Claude-compatible chat completion with support for text, images, PDFs, tools, and extended thinking.
Input parameters
Unique identifier for the end-user (max 128 characters).
Model identifier to use for generation. Example: 'anthropic/claude-4.5-sonnet-20250929'.
List of tools available for the model to call.
Top-k sampling parameter.
Nucleus sampling probability between 0 and 1.
List of fallback models to use if primary model fails.
Whether to stream the response. Default false.
System message to guide model behavior.
Plugins to enable for this request.
App display name for OpenRouter dashboard.
List of conversation messages. Each message has a role ('user' or 'assistant') and content.
Metadata for the request.
Provider routing preferences.
Configuration for extended thinking.
Maximum number of tokens to generate in the response.
Unique identifier for grouping related requests (max 128 characters).
Sampling temperature between 0 and 2.
Tool choice configuration.
App URL identifier used for tracking API usage per application.
Service tier for the request.
Configuration for controlling output behavior.
Sequences where generation should stop.
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
OPENROUTER_GET_CREDITSTool to get the current API credit balance for the authenticated user. Use before large or batch jobs to verify sufficient balance. A successful response may return total_credits=0, which confirms authentication but will cause all paid model generations to fail. Avoid polling this endpoint; call only as needed.
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
OPENROUTER_GET_CURRENT_KEYTool to get information about the currently authenticated API key. Use to check usage limits, spending, and key metadata.
Input parameters
The app display name allows you to customize how your app appears in OpenRouter's dashboard.
The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.
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
OPENROUTER_GET_GENERATIONTool to retrieve a generation result by its unique ID. Use after a generation completes to fetch metadata like token counts, cost, and latency.
Input parameters
Unique identifier of the generation to retrieve.
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
OPENROUTER_GET_MODELS_COUNTTool to get the total count of available models on OpenRouter. Use when you need to know how many models are available without fetching the full list.
Input parameters
The app display name allows you to customize how your app appears in OpenRouter's dashboard.
The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.
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
OPENROUTER_LIST_AVAILABLE_MODELSTool to list available models via OpenRouter API. Use after confirming authentication to fetch the model catalog. Use exact model IDs returned here in OPENROUTER_CREATE_CHAT_COMPLETION or OPENROUTER_CREATE_COMPLETION calls — hard-coded IDs may break when the catalog changes. Use exact author and slug values from this response as inputs to OPENROUTER_LIST_MODEL_ENDPOINTS. Models have varying capabilities (e.g., tools, reasoning); verify individual model capabilities before downstream use. Pricing and latency metadata may be null or approximate — handle missing values in routing logic.
Input parameters
Return RSS XML feed instead of JSON (BETA).
Valid categories for filtering models.
Use chat URLs instead of model page URLs in RSS items (only applies when use_rss=true) (BETA).
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
OPENROUTER_LIST_EMBEDDING_MODELSTool to list all available embeddings models via OpenRouter API. Returns a list of embeddings models with their properties including architecture, pricing, and capabilities.
Input parameters
The app display name allows you to customize how your app appears in OpenRouter's dashboard.
The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.
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
OPENROUTER_LIST_MODEL_ENDPOINTSTool to list endpoints for a specific model. Use after specifying model author and slug to get endpoint details including pricing, context length, and supported parameters. Some metadata fields (e.g., latency, pricing) may be null or approximate; handle missing values in routing logic.
Input parameters
Model's unique slug under the author namespace (e.g., 'gpt4').
Model author's username as it appears in the model URL (e.g., 'anon'). Use exact value from OPENROUTER_LIST_AVAILABLE_MODELS — approximate or incorrect values return empty results.
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
OPENROUTER_LIST_PROVIDERSTool to list all AI model providers available through the OpenRouter API. Use after authentication to retrieve available provider options for routing configuration. Providers differ in latency, context window sizes, and rate limits — switching providers affects these constraints. Newly added providers may not appear immediately due to catalog propagation delays.
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
OPENROUTER_LIST_USER_MODELSTool to list models filtered by user provider preferences, privacy settings, and guardrails. Use after authenticating to get models tailored to the user's configuration.
Input parameters
The app display name allows you to customize how your app appears in OpenRouter's dashboard.
The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.
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
OPENROUTER_LIST_ZDR_ENDPOINTSTool to preview the impact of Zero Data Retention (ZDR) on the available endpoints. Use to see which model endpoints remain accessible when ZDR is enabled.
Input parameters
The app display name allows you to customize how your app appears in OpenRouter's dashboard.
The app identifier should be your app's URL and is used as the primary identifier for rankings. This is used to track API usage per application.
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 55 agents privately built on Nagent that already use Openrouter.
Build on Nagent
Connect Openrouter 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 Openrouter, and click "Connect Now." You'll authenticate with an API key — Nagent handles credential storage and refresh automatically. Once connected, Openrouter is available to any agent in your workspace.
No. Nagent provides no-code integration for every tool. Once Openrouter is connected, you configure its 13 actions directly in the agent builder UI — no API calls, no boilerplate, no schema management.
Helix — Nagent's agentic agent builder — lets you drop Openrouter 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 Openrouter event fires, the agent kicks off automatically.
Every Openrouter 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 Openrouter ships with 13 pre-built artificial intelligence actions, you can layer custom logic around them inside Helix — pre/post-processing steps, conditional branches, retries, or stitching Openrouter together with other connected tools. For deeper customization, talk to our team about Nagent's Agentic AI Lab — forward-deployed engineers who build Openrouter-based workflows tailored to your business.