gte-base-en-v1.5

gte-base-en-v1.5

• Alibaba-NLP/gte-base-en-v1.5 ONNX format model

  • Language: english
  • Model Size: 137M parameters
  • Max Seq. Length: 8192
  • Dimension: 768
  • MTEB-en: 64.11
  • LoCo: 87.44

• Opensearch Model Config

  • model_type (architectures): BertModel
    • Original value: NewModel
  • framework_type: sentence_transformers
  • embedding_dimension: 768
  • all_config: can get from config.json

Download Url: https://github.com/magda-io/embedding_models/releases/download/gte-base-en-v1.5/model.zip
File SHA-256 hash: 55dfd0ae0b9140cf091a3ee26fe0542e00807c3827687ac2e66da27c12f01084
Size: 356121727 bytes
License: apache-2.0

gte-large-en-v1.5

gte-large-en-v1.5

This model currently doesn't work on OpenSearch v2.15.0 with the deploy error: "input mismatch, looking for: [input_ids, attention_mask, token_type_ids]".

• Alibaba-NLP/gte-large-en-v1.5 ONNX format model

  • Language: english
  • Model Size: 434M parameters
  • Max Seq. Length: 8192
  • Dimension: 1024
  • MTEB-en: 65.39
  • LoCo: 86.71

• Opensearch Model Config

  • model_type (architectures): BertModel
    • Original value: NewModel
  • framework_type: sentence_transformers
  • embedding_dimension: 1024
  • all_config: can get from config.json

Download Url: https://github.com/magda-io/embedding_models/releases/download/gte-large-en-v1.5/model.zip
File SHA-256 hash: 72e4037cd3e2a2499a3c92759159982982871a40715ca7c38f583934aa17306b
Size: 1271086226 bytes
License: apache-2.0