Error 500: "Model does not exist or you do not have access to it" during Vector Store insertion

[yunusgndzz571-ctrl]

Hi everyone,

I am encountering a 500 error within Flowise when trying to process documents using the Document Store.

The Issue:
When the system attempts to insert data into the vector store (documentStoreServices.insertIntoVectorStore), it fails with a 404 error regarding a specific model.

Error Log:

Context:

• It seems like the system is looking for a model named Chat-chroma-chunc-pluskim.
• The troubleshooting link provided in the error points to LangChain's MODEL_NOT_FOUND page.

Has anyone experienced this issue or knows where this specific model name might be coming from? Is this a configuration mismatch?

Any help would be appreciated. Thanks!

[xXMrNidaXx]

The model name "Chat-chroma-chunc-pluskim" looks like a corrupted or misconfigured model reference.

Root cause:
The vector store insertion is looking for an embedding model, but the configured model name is invalid.

Fixes:

1. Check your embedding model config
In the Document Store node:

• Embedding Model should be something like text-embedding-3-small (OpenAI) or nomic-embed-text (local)
• NOT a chatflow name or random string

2. Verify credentials

Settings → Credentials → OpenAI API (or your provider)

Ensure the API key has access to embedding models.

3. Clear corrupted config

• Delete the Document Store
• Create a new one with correct embedding model
• Re-upload documents

4. Check environment variables

# If self-hosted, verify:
OPENAI_API_KEY=sk-...

5. Model name source
The weird name suggests either:

• Copy-paste error in config
• Database corruption
• Variable substitution gone wrong

Debug:

# In Flowise logs, look for:
"embeddings" or "model" in the request payload

Quick fix:

1. Export your chatflow as JSON
2. Search for Chat-chroma-chunc-pluskim
3. Replace with correct model name
4. Re-import

We troubleshoot Flowise deployments at Revolution AI — this is usually a config mismatch in the embedding model field.