> ## Documentation Index
> Fetch the complete documentation index at: https://composio-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Eden AI

> Eden AI provides one API for accessing and orchestrating AI models across multiple providers for text, image, audio, document, and multimodal workloads.

![Eden AI logo](https://mintcdn.com/composio-preview/r7CO9m4_UWMVFnU7/images/toolkits/eden_ai.svg?fit=max\&auto=format\&n=r7CO9m4_UWMVFnU7\&q=85\&s=1ca69931e2540245b01b398fbd9aeb27)

* **Category:** artificial intelligence
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 10
* **Triggers:** 0
* **Slug:** `EDEN_AI`
* **Version:** 20261101\_00

## Tools

> [accordion] Count Anthropic Tokens
>
> **Slug:** `EDEN_AI_COUNT_ANTHROPIC_TOKENS`
>
> Count input tokens for an Anthropic-format message request before generation, including system content and tool definitions.
>
> #### Input Parameters
>
> **model** `string` *required*
>
> Anthropic-compatible model ID whose tokenizer should be used.
>
> **tools** `array`
>
> Anthropic tool definitions included in the measured request.
>
> **system** `string`
>
> System text or text content blocks included in the token count.
>
> **messages** `array` *required*
>
> Ordered, non-empty Anthropic-format messages to measure.
>
> **tool\_choice** `object`
>
> Tool-selection configuration included in the measured request.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Anthropic Message
>
> **Slug:** `EDEN_AI_CREATE_ANTHROPIC_MESSAGE`
>
> Generate a non-streaming Anthropic-compatible message with content blocks, stop reason, token usage, provider metadata, and Eden AI cost. Each call runs a paid model and consumes Eden AI credits.
>
> #### Input Parameters
>
> **model** `string` *required*
>
> Anthropic-compatible model ID in 'provider/model' form.
>
> **top\_p** `number`
>
> Nucleus sampling probability from 0 through 1.
>
> **system** `string`
>
> System prompt as text or an ordered list of Anthropic content blocks.
>
> **messages** `array` *required*
>
> Ordered, non-empty Anthropic-format user and assistant conversation messages.
>
> **fallbacks** `array`
>
> Up to three fallback model IDs to try in order if the primary model fails.
>
> **max\_tokens** `integer`
>
> Maximum number of tokens to generate; must be at least 1.
>
> **temperature** `number`
>
> Sampling temperature from 0 through 2.
>
> **stop\_sequences** `array`
>
> Text sequences that stop generation when encountered.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Chat Completion
>
> **Slug:** `EDEN_AI_CREATE_CHAT_COMPLETION`
>
> Generate a non-streaming OpenAI-compatible chat completion from a text-only conversation using an Eden AI model. Returns choices, token usage, provider, and cost; calls consume Eden AI credits.
>
> #### Input Parameters
>
> **model** `string` *required*
>
> Current model ID in 'provider/model' form from DISCOVER\_CAPABILITIES, or '@edenai' for dynamic routing.
>
> **messages** `array` *required*
>
> Ordered, non-empty text conversation sent to the model.
>
> **fallbacks** `array`
>
> Up to three model IDs to try in order if the primary model fails.
>
> **max\_tokens** `integer`
>
> Maximum number of tokens to generate; must be at least 1.
>
> **temperature** `number`
>
> Sampling temperature from 0 through 2.
>
> **response\_format** `object`
>
> Provider-compatible response format configuration, including optional JSON Schema output.
>
> **reasoning\_effort** `string ("minimal" | "low" | "medium" | "high" | "max" | "xhigh" | "disable" | "none")`
>
> Reasoning effort for models that support configurable reasoning.
>
> **router\_candidates** `array`
>
> Candidate model IDs considered for dynamic routing; valid only when model is '@edenai'.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Embeddings
>
> **Slug:** `EDEN_AI_CREATE_EMBEDDINGS`
>
> Create embeddings for one text or a batch of texts and return vectors with model, usage, provider, and cost metadata. Each call runs a paid model and consumes Eden AI credits.
>
> #### Input Parameters
>
> **user** `string`
>
> Stable end-user identifier for provider abuse monitoring.
>
> **input** `string` *required*
>
> One non-empty text or a non-empty batch of up to 2048 texts to embed.
>
> **model** `string` *required*
>
> Embedding model ID in 'provider/model' form from Discover AI Capabilities.
>
> **dimensions** `integer`
>
> Positive reduced vector size for models that support configurable dimensions.
>
> **encoding\_format** `string ("float" | "base64")`
>
> Return each embedding as float values or a base64-encoded string.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Create Response
>
> **Slug:** `EDEN_AI_CREATE_RESPONSE`
>
> Create or continue a synchronous OpenAI Responses-compatible text response, stored by default for later retrieval, and return its ID, structured output, text, usage, provider, and cost.
>
> #### Input Parameters
>
> **input** `string`
>
> Text input for a new turn; may be omitted only when previous\_response\_id supplies conversation state.
>
> **model** `string` *required*
>
> Current model ID in 'provider/model' form from DISCOVER\_CAPABILITIES.
>
> **store** `boolean`
>
> Store the response so GET\_RESPONSE can retrieve it later.
>
> **fallbacks** `array`
>
> Up to three fallback model IDs tried in order.
>
> **temperature** `number`
>
> Sampling temperature from 0 through 2.
>
> **instructions** `string`
>
> System or developer instructions for this response; not inherited from the previous response.
>
> **max\_output\_tokens** `integer`
>
> Requested maximum output tokens; Eden AI or the provider may enforce a higher minimum.
>
> **previous\_response\_id** `string`
>
> Stored response ID to continue a server-managed conversation.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Discover AI Capabilities
>
> **Slug:** `EDEN_AI_DISCOVER_CAPABILITIES`
>
> Find current Eden AI models or inspect Universal AI feature schemas, providers, pricing, and regions before an execution call.
>
> #### Input Parameters
>
> **query** `string`
>
> Case-insensitive text matched against model IDs, model names, alias\_of, feature names, and subfeature names.
>
> **region** `string`
>
> Keep records available in this region code, such as 'eu'.
>
> **feature** `string`
>
> Universal AI feature such as 'text'; omit to list all feature groups.
>
> **provider** `string`
>
> Keep model or Universal AI provider records matching this provider name.
>
> **model\_type** `string ("all" | "embeddings" | "moderation")`
>
> Model catalog to search when discovery\_type is 'models'.
>
> **subfeature** `string`
>
> Universal AI subfeature such as 'moderation'; requires feature and returns its simplified runtime schema.
>
> **max\_results** `integer`
>
> Maximum matching records to return after local filtering (1-100).
>
> **discovery\_type** `string ("models" | "universal_ai")`
>
> Choose 'models' for executable model records or 'universal\_ai' for feature and input-schema discovery.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Cost Overview
>
> **Slug:** `EDEN_AI_GET_COST_OVERVIEW`
>
> Return current account credits and, when a date range is supplied, aggregated consumption filtered by provider or subfeature.
>
> #### Input Parameters
>
> **end\_date** `string`
>
> Last UTC calendar date to include (YYYY-MM-DD); requires start\_date.
>
> **provider** `string`
>
> Limit consumption data to one provider; requires a date range.
>
> **start\_date** `string`
>
> First UTC calendar date to include (YYYY-MM-DD); requires end\_date.
>
> **subfeature** `string`
>
> Limit consumption data to one Eden AI subfeature; requires a date range.
>
> **aggregation** `string ("daily" | "weekly" | "monthly" | "yearly")`
>
> Time bucket for consumption data when a date range is supplied.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Response
>
> **Slug:** `EDEN_AI_GET_RESPONSE`
>
> Retrieve a previously stored Responses API object by ID, including its status, text output, structured output, usage, and canonical model.
>
> #### Input Parameters
>
> **response\_id** `string` *required*
>
> Response ID returned by CREATE\_RESPONSE.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Moderate Text
>
> **Slug:** `EDEN_AI_MODERATE_TEXT`
>
> Classify one text or a batch of texts for safety categories and return flags, category scores, applied input types, provider, and cost. Each call runs a paid model and consumes Eden AI credits.
>
> #### Input Parameters
>
> **input** `string` *required*
>
> One text or a non-empty batch of texts to classify for safety.
>
> **model** `string`
>
> Moderation model ID in 'provider/model' form. Discover current alternatives before overriding this default.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Run Universal AI
>
> **Slug:** `EDEN_AI_RUN_UNIVERSAL_AI`
>
> Run one synchronous Eden AI Universal AI feature using a current model and feature-specific input. Call DISCOVER\_CAPABILITIES first for the exact model ID and runtime input schema. Returns normalized output, provider, and cost, with the raw provider response only when requested. Each call runs a paid model and consumes Eden AI credits.
>
> #### Input Parameters
>
> **input** `object` *required*
>
> Non-empty feature-specific input object. Copy the exact field names and value shapes returned by DISCOVER\_CAPABILITIES for the selected feature and subfeature; Eden AI validates this object dynamically at runtime.
>
> **model** `string` *required*
>
> Full model ID in 'feature/subfeature/provider\[/model]' form. Call DISCOVER\_CAPABILITIES immediately before this tool to obtain a current model ID and its exact input schema.
>
> **fallbacks** `array`
>
> Up to three fallback provider names, provider/model identifiers, or full Universal AI model IDs, tried in order.
>
> **provider\_params** `object`
>
> Provider-specific parameters documented in the discovered feature schema. Omit this object unless the selected provider requires or supports them.
>
> **show\_original\_response** `boolean`
>
> Include the raw provider response in addition to Eden AI's normalized output. Leave false for a smaller, provider-independent result.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not
