Skip to content

when collection set function with openai model, search with vector or text return error #500

@coder-van

Description

@coder-van

Describe the bug:

const client = await getMilvusClient();
const params: any = {
collection_name: this.collectionName,
data: embeddingsOrText,
limit: topN,
output_fields: ['primary_key', 'metadata', 'document', 'toolkit_symbol'],
...kwargs
}
console.log('Milvus search params:', JSON.stringify({...params, data: '[embedding]'}, null, 2));

when set search data vector return
Error: Milvus search error: UnexpectedError fail to search on QueryNode 2505: worker(2505) query failed: parser searchRequest failed: => vector dimension mismatch, expected vector size(byte) 4096, actual 12758. at /root/milvus/internal/core/src/query/Plan.cpp:91

when set search data text return expected vector size(byte) 4096, actual 5

when i use axios call api, search is success, than i use version 2.5.12, search is successed
Steps to reproduce:

  1. use 2.6.9 sdk
  2. use openai function
  3. try search

Milvus-node-sdk version:
2.6.9

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions