BAAI/bge-m3 · Discussions

#75 opened 11 months ago by

HonorOfKing

The GPU memory is not being released after inference?

#74 opened 11 months ago by

chf7410

Multilingual

#73 opened 11 months ago by

Sardar

1_Pooling 这个目录的作用是？

7

#72 opened 12 months ago by

Godsing

Adding `safetensors` variant of this model

#71 opened 12 months ago by

SFconvertbot

Transformers方式调用bge-m3是否有方法返回稀疏向量？

#70 opened 12 months ago by

imdoge

No sentence-transformers model found with name BAAI/bge-m3. Creating a new one with mean pooling.

5

#69 opened 12 months ago by

hhgz

How did you train m3-retromae?

#66 opened about 1 year ago by

hotchpotch

If deploy m3 and m3 raranker model on a same gpu service，will it improve the gpu efficiency in the concurrent request environment

#65 opened about 1 year ago by

seetimee

[AUTOMATED] Model Memory Requirements

🔥 4

#64 opened about 1 year ago by

model-sizer-bot

I have question on result.

#62 opened about 1 year ago by

Methodfunc

Model size

#60 opened about 1 year ago by

Juli784

Long context embedding & chunking on same document with bge-m3

#59 opened about 1 year ago by

DracoDev

Discussions about aggregate score computing (and question about banchmarks against colbert / splade)

#58 opened about 1 year ago by

bobox

Maximum and Optimal Text Size That Can Be Vectror Embedded with bge-m3?

#57 opened about 1 year ago by

DracoDev

Does bge-m3 understand html tages <h1>, <h2> , <code> etc?

8

#56 opened about 1 year ago by

DracoDev

Issue: Semantic Similarity Score

#55 opened about 1 year ago by

linhphanff

不同平台向量化后结果有差异

#54 opened about 1 year ago by

compute_score 问题

#53 opened about 1 year ago by

How to use the ONNX format of BERT-based model bge-m3, how to load the model and output sentence embeddings.

#52 opened about 1 year ago by

panjiayi

生产中使用API问题

#51 opened about 1 year ago by

如何运行模型

#49 opened about 1 year ago by

how to use trainer 'resume_from_checkpoint'?

#48 opened about 1 year ago by

COCO0414

Reduce dimension of colbert vector

#47 opened about 1 year ago by

feliciaAI

Adding `safetensors` variant of this model

#46 opened about 1 year ago by

SFconvertbot

How to enable fp16 option in fintuning?

9

#45 opened about 1 year ago by

comet24082002

Got this error: AttributeError: 'NoneType' object has no attribute 'size'

#43 opened about 1 year ago by

latuan

Is bge-m3 compatible with OpenCLIP?

#42 opened about 1 year ago by

seanchenupland

Colbert Mode Usage

#41 opened about 1 year ago by

pulkitchahar

Cuda out of memory

#40 opened about 1 year ago by

comet24082002

Evaluation for finetuned bge-m3 model

#39 opened about 1 year ago by

comet24082002

Adding `safetensors` variant of this model

#38 opened about 1 year ago by

SFconvertbot

Got this error: TypeError: SafeTensorsInfo.init() got an unexpected keyword argument 'sharded'

10

#37 opened about 1 year ago by

favioespinosav

Any 'small' version for this model with 384 dimensions ?

4

#36 opened about 1 year ago by

zaobao

Do I need to add the prefix "query: " and "passage: " to input texts?

5

#35 opened about 1 year ago by

zaobao

Endless loop with multiple GPU ?

#34 opened about 1 year ago by

Grosper

Using bge-m3 for clustering and search

#33 opened about 1 year ago by

talavivi

Will the component models be available separately?

#31 opened over 1 year ago by

libryo-ai

Reranker

4

#30 opened over 1 year ago by

Totole

list of languages supported

👍 2

#29 opened over 1 year ago by

andrew123456789

convert token to the original form

#28 opened over 1 year ago by

DAIEF

m3需要和base做merge吗

#27 opened over 1 year ago by

biaodiluer

Which matrix did you use for Benchmarks from the open-source community

#26 opened over 1 year ago by

DAIEF

支持langchain使用bge-m3模型吗？

8

#25 opened over 1 year ago by

Nicole828

This models is also best for Finnish in my comparison

👍 1

#24 opened over 1 year ago by

RASMUS

Code and some results for comparing with other embedding models on multilingual data

👍 2

#23 opened over 1 year ago by

Yannael

入参之间会影响embedding的结果吗

#22 opened over 1 year ago by

biaodiluer

关于多阶段训练

#21 opened over 1 year ago by

xxxcliu

bge-M3和baai_general_embedding是什么关系