gte-small

Fork of thenlper/gte-small with ONNX to work with Transformers.js.

---

gte-small

General Text Embeddings (GTE) model. Towards General Text Embeddings with Multi-stage Contrastive Learning

The GTE models are trained by Alibaba DAMO Academy. They are mainly based on the BERT framework and currently offer three different sizes of models, including GTE-large, GTE-base, and GTE-small. The GTE models are trained on a large-scale corpus of relevance text pairs, covering a wide range of domains and scenarios. This enables the GTE models to be applied to various downstream tasks of text embeddings, including information retrieval, semantic textual similarity, text reranking, etc.

Metrics

Performance of the GTE models compared with other popular text embedding models on the MTEB benchmark. For more detailed comparison results, please refer to the MTEB leaderboard.

Usage

Deno

import { env, pipeline } from "https://cdn.jsdelivr.net/npm/@xenova/[email protected]";

// Some config for Deno
env.useBrowserCache = false;
env.allowLocalModels = false;

// Give it any input you want
const input = "Hello AI";

// Create the pipeline
const pipe = await pipeline(
  "feature-extraction",
  "koxy-ai/gte-small"
);

// Generate the embedding
const output = await pipe(input, {
  pooling: "mean",
  normalize: true
});

// Extract the embedding from the output
const embedding = Array.from(output.data);

// Do anything with the embedding
console.log(embedding);

Browser

Using Javascript modules.

<script type="module">

import { pipeline } from "https://cdn.jsdelivr.net/npm/@xenova/[email protected]";

// Create the pipeline
const setPipe = async () => {
  return await pipeline(
    "feature-extraction",
    "koxy-ai/gte-small"
  );
};

const generateEmbedding = async (input) => {
  const pipe = await setPipe();
  const output = await pipe(input, {
    pooling: "mean",
    normalize: true
  });
  return Array.from(output.data);
};

export default generateEmbedding;

</script>

Node JS

npm i @xenova/transformers

import { pipeline } from "@xenova/transformers";

(async () => {
  // Give it any input you want
  const input = "Hello AI";

  // Create the pipeline
  const pipe = await pipeline(
    "feature-extraction",
    "koxy-ai/gte-small"
  );

  // Generate the embedding
  const output = await pipe(input, {
    pooling: "mean",
    normalize: true
  });

  // Extract the embedding from the output
  const embedding = Array.from(output.data);

  // Do anything with the embedding
  console.log(embedding);
})();

Limitation

This model exclusively caters to English texts, and any lengthy texts will be truncated to a maximum of 512 tokens.

Citation

@misc{li2023general,
      title={Towards General Text Embeddings with Multi-stage Contrastive Learning}, 
      author={Zehan Li and Xin Zhang and Yanzhao Zhang and Dingkun Long and Pengjun Xie and Meishan Zhang},
      year={2023},
      eprint={2308.03281},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}