FlowiseAI/Flowise · error · Error

There must be a LLM model connected to LLM Filter Retriever

Error message

There must be a LLM model connected to LLM Filter Retriever

What it means

Thrown by LLMFilterCompressionRetriever.init when nodeData.inputs.model is falsy. LLMChainExtractor.fromLLM requires a language model to decide which documents to keep; with no model wired the node cannot build the compressor and aborts.

Source

Thrown at packages/components/nodes/retrievers/LLMFilterRetriever/LLMFilterCompressionRetriever.ts:78

                description: 'Array of document objects containing metadata and pageContent',
                baseClasses: ['Document', 'json']
            },
            {
                label: 'Text',
                name: 'text',
                description: 'Concatenated string from pageContent of documents',
                baseClasses: ['string', 'json']
            }
        ]
    }

    async init(nodeData: INodeData, input: string): Promise<any> {
        const baseRetriever = nodeData.inputs?.baseRetriever as BaseRetriever
        const model = nodeData.inputs?.model as BaseLanguageModel
        const query = nodeData.inputs?.query as string
        const output = nodeData.outputs?.output as string

        if (!model) throw new Error('There must be a LLM model connected to LLM Filter Retriever')

        const retriever = new ContextualCompressionRetriever({
            baseCompressor: LLMChainExtractor.fromLLM(model),
            baseRetriever: baseRetriever
        })

        if (output === 'retriever') return retriever
        else if (output === 'document') return await retriever._getRelevantDocuments(query ? query : input)
        else if (output === 'text') {
            let finaltext = ''

            const docs = await retriever._getRelevantDocuments(query ? query : input)

            for (const doc of docs) finaltext += `${doc.pageContent}\n`

            return handleEscapeCharacters(finaltext, false)
        }

View on GitHub (pinned to abe4a8601a)

Solutions

Connect a Chat LLM or LLM node to the LLM Filter Retriever's model input.
Confirm the connected model node initializes successfully on its own.
Ensure the model output type is BaseLanguageModel (not Embeddings).

Example fix

// before: model input unconnected
// after: wire a ChatOpenAI / ChatAnthropic (etc.) node into the model input

Defensive patterns

Strategy: type-guard

Validate before calling

import { BaseLanguageModel } from '@langchain/core/language_models/base'
function ensureModel(model: unknown): BaseLanguageModel {
  if (!model) throw new Error('Connect an LLM to the LLM Filter Retriever model input.')
  if (typeof (model as any)._generate !== 'function' && typeof (model as any).invoke !== 'function') {
    throw new Error('Connected node is not a BaseLanguageModel.')
  }
  return model as BaseLanguageModel
}

Type guard

function isBaseLanguageModel(m: unknown): m is BaseLanguageModel {
  return !!m && typeof m === 'object' && ('_generate' in m || 'invoke' in m) && '_modelType' in (m as any)
}

Try / catch

try {
  await retriever.init(nodeData, input)
} catch (e) {
  if (e instanceof Error && /LLM model connected/.test(e.message)) {
    // highlight the model input port
  }
  throw e
}

Prevention

Validate the chatflow graph: every LLM Filter Retriever has an incoming model edge.
Use a type guard on nodeData.inputs.model before init.
Test that the connected model node initializes in isolation.

When it happens

Trigger: LLM Filter Retriever node has no model input connected; the connected model node errored on init and produced undefined; the wrong input type is wired (e.g. an embedding model where a chat model is expected).

Common situations: Forgotten edge in the chatflow graph; model node deleted after wiring; mismatch between the expected BaseLanguageModel input and what is connected.

Related errors

AI-assisted analysis of FlowiseAI/Flowise@abe4a8601a (2026-08-12). Data as JSON: /api/errors/333e1d870af6adbf. Report an issue: GitHub.