# Cohere integration

> Enterprise AI model API: chat (Command models), embeddings, rerank, tokenization, datasets, embed jobs, batches, and transcription.

- Authentication: apiKey

## Tools (26)

- **Cancel a batch**: Cancels an in-progress batch.
- **Cancel an Embed Job**: Cancels an active embed job.
- **Chat API (v1)**: Legacy v1 chat: generates a text response to a user message, with RAG, connectors and tools.
- **Chat API (v2)**: Generates a text response from a list of chat messages, with tools, documents with citations, structured output and streaming.
- **Check API key**: Checks that the API key in the Authorization header is valid and active.
- **Create a batch**: Creates and runs a batch job over an uploaded dataset of requests.
- **Create a Dataset**: Creates a dataset by uploading a file.
- **Create an Embed Job**: Launches an async embed job over an embed-input dataset.
- **Create a transcription**: Transcribes an audio file to text.
- **Delete a Dataset**: Deletes a dataset by ID.
- **Detokenize**: Converts tokens back to text.
- **Embed API (v1)**: Legacy v1 embed: returns text and image embeddings.
- **Embed API (v2)**: Returns text, image and mixed-content embeddings for search, clustering and classification.
- **Fetch an Embed Job**: Retrieves details and status of an embed job.
- **Get a Dataset**: Retrieves a dataset by ID.
- **Get a Model**: Returns details of a model by name.
- **Get Dataset Usage**: Returns dataset storage usage for the organization.
- **List batches**: Lists batches for the current user.
- **List Datasets**: Lists datasets that have been created.
- **List Embed Jobs**: Lists all embed jobs for the user.
- **List Models**: Lists models available to the account.
- **Parse API (v2)**: Parses a document image into structured markdown or blocks.
- **Rerank API (v1)**: Legacy v1 rerank: orders documents by relevance to a query.
- **Rerank API (v2)**: Orders a list of texts by relevance to a query.
- **Retrieve a batch**: Retrieves a batch and its status.
- **Tokenize**: Splits text into tokens using a model's byte-pair encoding tokenizer.

Connect Cohere in General Input: https://www.generalinput.com/apps/cohere