ActionCohereUpdated September 2026

How do I send a message to Cohere's Command model?

Short answer: You can chat in Cohere by hand from its own interface, but it won’t repeat itself. On TinyCommand, add the Cohere Chat action to a workflow, map its 4 inputs from any upstream app, and it runs automatically every time the trigger fires. No code, and a free tier to start.

Chat in Cohere — start free All 4 Cohere actions

Chat in Cohere — start free

Inputs

The fields this action accepts.

Every field can be mapped from an upstream trigger, AI step, table row, or hard-coded literal.

Field	Type	Required	Description
Model model	options	Required	Model. Options: Command R+, Command R, Command, Command Light
Message message	string	Required	Message. Example: Summarize this quarterly report...
System Preamble preamble	string	Optional	Instructions for the model
Temperature temperature	string	Optional	Temperature. Example: 0.3

Sample request

{
  "model": "{{trigger.model}}",
  "message": "e.g. Summarize this quarterly report...",
  "preamble": "{{trigger.preamble}}",
  "temperature": "e.g. 0.3"
}

Returns

{
  "id": "abc123",
  "usage": {
    "input_tokens": 50,
    "output_tokens": 200
  },
  "message": {
    "role": "assistant",
    "content": "Based on the report..."
  },
  "finish_reason": "COMPLETE"
}

Use these fields in downstream nodes for routing, logging, or error handling.

Triggered by

Apps that pair well as the trigger for Chat.

Any of these apps can fire this action as part of a workflow.

Google Sheets → Cohere

2 Google Sheets triggers

HubSpot → Cohere

18 HubSpot triggers

FAQ

Questions about Chat.

What does the Chat action do in Cohere?

Sends a message and gets a response from one of Cohere's Command models. Cohere is known for enterprise-grade chat with strong RAG and grounded-answer features.

What inputs does Chat require?

Required: Model, Message. Every input accepts a static value or a variable from any upstream node in your workflow.

Can I use dynamic inputs from earlier workflow nodes?

Yes. Any field on this action can pull values from upstream nodes, whether that's a form response, a trigger payload, an AI output, or a lookup result.

What happens if Cohere returns an error?

The run history shows the failed step with its input and the error message Cohere returned, so you can fix the input and run the workflow again. Automatic retries are built into HTTP steps only, where you can also set your own retry count and delay.

Does Chat support batch operations?

Run Chat inside a loop to process each item in an array. TinyCommand does not throttle calls for you, so for large arrays add a Delay step if you need to stay within Cohere's rate limits.

More actions

Other Cohere actions.

Action

Chat with Tools

Chats with Cohere Command models with tool/function calling enabled. Use to let the model fetch data, run actions, or build agents within Cohere's ecosystem.

Action

Create Embeddings

Generates vector embeddings for text using Cohere's embed-v3 model. Supports input_type (search_document, search_query, classification, clustering), which materially improves retrieval quality, so set it correctly.

Action

Rerank

Reranks a list of documents by relevance to a query using Cohere's Rerank model. Essential last-mile step in any serious RAG pipeline; meaningfully lifts precision over vector-only retrieval.

Chat in Cohere — start free