Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

BGE-Large-en-v1.5

Model basics

Reasoning traces

Not supported

Context length

512 tokens

Max output length

No data

Model type

embedding模型

Release date

2023-09-12

Model file size

1.34GB

MoE architecture

Total params / Active params

3.26B / N/A

Knowledge cutoff

No data

Inference modes

No mode data

BGE-Large-en-v1.5

Open source & experience

Code license

MIT License

Weights license

MIT License- 免费商用授权

GitHub repo

https://github.com/FlagOpen/FlagEmbedding

Hugging Face

https://huggingface.co/BAAI/bge-large-en-v1.5

Live demo

No live demo

BGE-Large-en-v1.5

Official resources

Paper

C-Pack: Packaged Resources To Advance General Chinese Embedding

DataLearnerAI blog

No blog post yet

BGE-Large-en-v1.5

API details

API speed

No data

No public API pricing yet.

BGE-Large-en-v1.5

Benchmark Results

No benchmark data to show.

BGE-Large-en-v1.5

Publisher

北京智源人工智能研究院

View publisher details

BAAI General Embedding - Large-EN-1.5

Model Overview

北京智源人工智能研究院开源的向量大模型，bge-large-en-v1.5是其参数规模最大的一个，有3.26亿参数，英文版。完全开源。在MTEB和C-MTEB上得分都非常高。

1.5版本的bge-large-en-v1.5主要减轻了相似性分布问题，并加强了无需指令的检索能力。

该模型最高输入512个tokens，输出的向量维度是1024。

DataLearner 官方微信

欢迎关注 DataLearner 官方微信，获得最新 AI 技术推送