FlowiseAI/Flowise · error · Error

Collection ${args.collectionName} not found. You can create

Error message

Collection ${args.collectionName} not found. You can create a new Collection by providing embeddingDimensions.

What it means

Thrown by createNewCollection() when the named collection does not exist on the Zep server and args.embeddingDimensions is not supplied. Because isAutoEmbedded is false (the store receives precomputed embeddings from Flowise's embeddings model), Zep cannot size the collection itself and requires embeddingDimensions at creation time.

Source

Thrown at packages/components/nodes/vectorstores/Zep/Zep.ts:223

    async initializeCollection(args: IZepConfig & Partial<ZepFilter>) {
        this.client = await ZepClient.init(args.apiUrl, args.apiKey)
        try {
            this.collection = await this.client.document.getCollection(args.collectionName)
        } catch (err) {
            if (err instanceof Error) {
                if (err.name === 'NotFoundError') {
                    await this.createNewCollection(args)
                } else {
                    throw err
                }
            }
        }
    }

    async createNewCollection(args: IZepConfig & Partial<ZepFilter>) {
        if (!args.embeddingDimensions) {
            throw new Error(
                `Collection ${args.collectionName} not found. You can create a new Collection by providing embeddingDimensions.`
            )
        }

        this.collection = await this.client.document.addCollection({
            name: args.collectionName,
            description: args.description,
            metadata: args.metadata,
            embeddingDimensions: args.embeddingDimensions,
            isAutoEmbedded: false
        })
    }

    async similaritySearchVectorWithScore(
        query: number[],
        k: number,
        filter?: Record<string, unknown> | undefined
    ): Promise<[Document, number][]> {

View on GitHub (pinned to abe4a8601a)

Solutions

1. Set the Zep node's 'dimension' input to match your embeddings model (e.g. 1536 for OpenAI text-embedding-ada-002, 1536/3072 for text-embedding-3-*, etc.).
2. Or pre-create the collection on the Zep server with the correct embeddingDimensions.
3. Double-check the collection name spelling — if it was meant to match an existing collection, correct it.

Example fix

// before: dimension not set on node
nodeData.inputs?.dimension // undefined -> createNewCollection throws

// after
nodeData.inputs.dimension = 1536 // match embeddings model

// Ensure embeddingDimensions is set before the store tries to create the collection
function ensureZepDimension(args) {
    if (!args.embeddingDimensions || args.embeddingDimensions <= 0) {
        throw new Error(
            `Collection ${args.collectionName} does not exist. Set embeddingDimensions (e.g. 1536 for OpenAI ada-002) so it can be created.`
        )
    }
}

function hasEmbeddingDimensions(args) {
    return typeof args.embeddingDimensions === 'number' && args.embeddingDimensions > 0
}

try {
    await store.initializeCollection(args)
} catch (e) {
    if (/not found.*embeddingDimensions/i.test(e.message)) {
        // set dimension on the node and retry
    }
    throw e
}

When it happens

Trigger: First use of a brand-new collection name while nodeData.inputs.dimension is undefined/0/falsy, so args.embeddingDimensions is missing when the store tries to create the collection.

Common situations: Forgot to set the 'dimension' input on the Zep node; collection name typo (intended to reference an existing collection but actually new); Zep server data was wiped/reset so the expected collection no longer exists.

Related errors

• cannot provide both `filter` and `this.filter`
• Model is required
• Model is required
• Scenarios are required
• Invalid JSON in executeFlowOverrideConfig: ${parseError.mess

AI-assisted analysis of FlowiseAI/Flowise@abe4a8601a (2026-08-12). Data as JSON: /api/errors/81155468b7505287. Report an issue: GitHub.