Late Chunking support for Jina Embeddings (#4002)

* Late Chunking support for Jina Embeddings

* improved constructor handling. remove redundant code.

* remove unnessary function overrides, since late_chunking parameter only needs to be stored in the class.
This commit is contained in:
Gene Ruebsamen
2025-02-08 06:10:18 -08:00
committed by GitHub
parent 96dd1aaeea
commit 0449e80395
@@ -2,6 +2,16 @@ import { ICommonObject, INode, INodeData, INodeParams } from '../../../src/Inter
import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils' import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
import { JinaEmbeddings } from '@langchain/community/embeddings/jina' import { JinaEmbeddings } from '@langchain/community/embeddings/jina'
class ExtendedJinaEmbeddings extends JinaEmbeddings {
private late_chunking: boolean
constructor(fields: ConstructorParameters<typeof JinaEmbeddings>[0] & { late_chunking?: boolean }) {
const { late_chunking = false, ...restFields } = fields
super(restFields)
this.late_chunking = late_chunking
}
}
class JinaAIEmbedding_Embeddings implements INode { class JinaAIEmbedding_Embeddings implements INode {
label: string label: string
name: string name: string
@@ -17,7 +27,7 @@ class JinaAIEmbedding_Embeddings implements INode {
constructor() { constructor() {
this.label = 'Jina Embeddings' this.label = 'Jina Embeddings'
this.name = 'jinaEmbeddings' this.name = 'jinaEmbeddings'
this.version = 2.0 this.version = 3.0
this.type = 'JinaEmbeddings' this.type = 'JinaEmbeddings'
this.icon = 'JinaAIEmbedding.svg' this.icon = 'JinaAIEmbedding.svg'
this.category = 'Embeddings' this.category = 'Embeddings'
@@ -34,7 +44,7 @@ class JinaAIEmbedding_Embeddings implements INode {
label: 'Model Name', label: 'Model Name',
name: 'modelName', name: 'modelName',
type: 'string', type: 'string',
default: 'jina-embeddings-v2-base-en', default: 'jina-embeddings-v3',
description: 'Refer to <a href="https://jina.ai/embeddings/" target="_blank">JinaAI documentation</a> for available models' description: 'Refer to <a href="https://jina.ai/embeddings/" target="_blank">JinaAI documentation</a> for available models'
}, },
{ {
@@ -44,6 +54,15 @@ class JinaAIEmbedding_Embeddings implements INode {
default: 1024, default: 1024,
description: description:
'Refer to <a href="https://jina.ai/embeddings/" target="_blank">JinaAI documentation</a> for available dimensions' 'Refer to <a href="https://jina.ai/embeddings/" target="_blank">JinaAI documentation</a> for available dimensions'
},
{
label: 'Allow Late Chunking',
name: 'allowLateChunking',
type: 'boolean',
description:
'Refer to <a href="https://jina.ai/embeddings/" target="_blank">JinaAI documentation</a> guidance on late chunking',
default: false,
optional: true
} }
] ]
} }
@@ -51,13 +70,15 @@ class JinaAIEmbedding_Embeddings implements INode {
async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> { async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
const modelName = nodeData.inputs?.modelName as string const modelName = nodeData.inputs?.modelName as string
const modelDimensions = nodeData.inputs?.modelDimensions as number const modelDimensions = nodeData.inputs?.modelDimensions as number
const allowLateChunking = nodeData.inputs?.modelDimensions as boolean
const credentialData = await getCredentialData(nodeData.credential ?? '', options) const credentialData = await getCredentialData(nodeData.credential ?? '', options)
const apiKey = getCredentialParam('jinaAIAPIKey', credentialData, nodeData) const apiKey = getCredentialParam('jinaAIAPIKey', credentialData, nodeData)
const model = new JinaEmbeddings({ const model = new ExtendedJinaEmbeddings({
apiKey: apiKey, apiKey: apiKey,
model: modelName, model: modelName,
dimensions: modelDimensions dimensions: modelDimensions,
late_chunking: allowLateChunking
}) })
return model return model