n8n Embeddings Mistral Cloud nodeConfigure Embeddings Mistral Cloud in n8n.
Embeddings turn your text into numbers a vector store can search. The n8n Embeddings Mistral Cloud node does that job with a Mistral model, through 2 parameters: Model and Options. It is a sub-node, so it runs only when a root node asks it for vectors.
Verified Trustpilot reviews · AI, automation & growth agency
What does the n8n Embeddings Mistral Cloud node do?
It computes embeddings, numerical representations of text as vectors, so AI can compare pieces of content by meaning instead of by keyword. In n8n it never sits alone on the canvas. You attach it to the Embeddings port of a root node, a vector store or a chain, and that root node hands it the text to convert.
Three shapes of workflow cover most of what gets built with it. The first is indexing: a document comes in, a text splitter cuts it into chunks, the vector store calls this node for each chunk and writes the vectors. The second is retrieval: a question arrives, the same node converts that question, and the store returns the closest chunks. The third is a question-answering flow where both sides happen in one run, which is exactly the pattern of the public templates around Qdrant and Mistral (2339, 2335, 2341).
The rule that saves you debugging time: the model that indexed your data is the model that has to query it. Vectors written with one model are not comparable with vectors from another, so switching Model after an index exists means reindexing everything. Pick it once, write it down somewhere.
When do you reach for something else? If your root node already carries its own embeddings provider and you have no Mistral credential, use that provider instead. If you only need a chat answer with no document search behind it, you need a chat model sub-node, not this one, and something like OpenAI covers that. And if the goal is an assistant that calls other systems rather than reading documents, the work belongs to AI Agent with tools such as Call n8n Workflow Tool.
The limits are worth knowing before you build. This node exposes exactly 2 parameters, and neither of them controls the vector store, the chunking or the retrieval count: those live on the nodes around it. It has no main input, so you cannot test it by clicking it on its own, and version 1 is the only version in the catalog. Mistral bills its own API calls under its own terms; n8n adds nothing on top.
If you want a broader picture of where n8n fits before committing an AI pipeline to it, the n8n review goes through the platform itself, and n8n training covers building these flows hands-on.
What credential does the node need?
- 01
Create the Mistral credential in n8n
Open the Credentials menu in n8n and add the Mistral credential, then paste the API key from your Mistral Cloud account. The credential is stored once and reused by every workflow on the instance, so a second embedding node or a Mistral chat model later on picks the same entry from the dropdown instead of asking for the key again.
- 02
Attach the credential to the node
Drop the node on the canvas from the Embeddings port of your root node, then select the credential you just created in its credential field. Nothing else on this node is authentication related: there is no Authentication selector to choose between methods, the Mistral credential is the only way in.
- 03
Run the parent node to test it
Because this is a sub-node with no main input, the play button that matters is the one on the root node. Execute the vector store or the chain it feeds, and watch that run: if the key is wrong or missing, the failure surfaces there, on the root node, not on a node that never ran by itself.
Where does the node plug in?
n8n splits AI work between a root node, which receives the workflow items, and sub-nodes clipped onto its ports, one type of sub-node per port. This node is a sub-node on the embeddings side.
Output (what it plugs into)
- Embeddings
ai_embedding
01Embeddings
This is the only connection the node has, and it is an output: the node offers embeddings, a root node consumes them. Without that link the node sits inert on the canvas, since a sub-node never executes on its own.
Key parameters
- Required: the node has no main input, so the Embeddings port is what makes it run at all.
How do you set its parameters?
The Embeddings Mistral Cloud node has 2 parameters. For each one: the node as you configure it in n8n, what the parameter changes, and our field notes.
Model
modelWhat you see in n8n
Notes & use cases
Picks which Mistral model computes the embeddings. It is a dropdown, and the choice decides the shape of the vectors your store will hold, which is why it is the one setting you lock before indexing anything.
Key parameters
- Model: the model which will compute the embeddings, selected from the options list rather than typed in.
Options
optionsWhat you see in n8n
Notes & use cases
A collection you open with Add Option, empty until you need it. Two extra settings live inside, both about how text reaches Mistral rather than about the model itself.
Key parameters
- Batch Size: the maximum number of documents sent in each request, useful when a splitter hands over a long queue of chunks at once.
- Strip New Lines: removes new line characters from the input text when turned on, which n8n does by default, and leaves them in place when turned off.
Need help automating Embeddings Mistral Cloud with n8n?
A person reads every message.
Questions people ask next
01Is the Embeddings Mistral Cloud node included in n8n?
02What do you need for it to work?
03What are its limits?
04What do you plug into the n8n Embeddings Mistral Cloud node to start?
05n8n or Make for this kind of AI workflow?
Get our weekly integration tips.
No spam. Unsubscribe anytime.
