Top AI lists by month and monthly visits.
Top AI lists by category and monthly visits.
Top AI lists by region and monthly visits.
Top AI lists by source and monthly visits.
Top AI lists by revenue and real traffic.
Fork of https://huggingface.co/BAAI/bge-small-en with ONNX weights to be compatible with Transformers.js. See JavaScript usage .
Model List | Usage | Evaluation | Train | License
For more details please refer to our GitHub repo: FlagEmbedding .
FlagEmbedding can map any text to a low-dimensional dense vector which can be used for tasks like retrieval, classification, clustering, or semantic search. And it also can be used in vector databases for LLMs.
************* 🌟 Updates 🌟 *************
bge-large-*
(short for BAAI General Embedding) Models,
rank 1st on MTEB and C-MTEB benchmark!
:tada: :tada:
bge
is short for
BAAI general embedding
.
Model | Language | Description | query instruction for retrieval |
---|---|---|---|
BAAI/bge-large-en | English | :trophy: rank 1st in MTEB leaderboard |
Represent this sentence for searching relevant passages:
|
BAAI/bge-base-en | English | rank 2nd in MTEB leaderboard |
Represent this sentence for searching relevant passages:
|
BAAI/bge-small-en | English | a small-scale model but with competitive performance |
Represent this sentence for searching relevant passages:
|
BAAI/bge-small-en | Chinese | :trophy: rank 1st in C-MTEB benchmark |
为这个句子生成表示以用于检索相关文章:
|
BAAI/bge-small-en-noinstruct | Chinese | This model is trained without instruction, and rank 2nd in C-MTEB benchmark | |
BAAI/bge-base-zh | Chinese |
a base-scale model but has similar ability with
bge-large-zh
|
为这个句子生成表示以用于检索相关文章:
|
BAAI/bge-small-zh | Chinese | a small-scale model but with competitive performance |
为这个句子生成表示以用于检索相关文章:
|
This model can be used with both Python and JavaScript .
pip install -U FlagEmbedding
See FlagEmbedding for more methods to install FlagEmbedding.
from FlagEmbedding import FlagModel
sentences = ["样例数据-1", "样例数据-2"]
model = FlagModel('Supabase/bge-small-en', query_instruction_for_retrieval="为这个句子生成表示以用于检索相关文章:")
embeddings = model.encode(sentences)
print(embeddings)
# for retrieval task, please use encode_queries() which will automatically add the instruction to each query
# corpus in retrieval task can still use encode() or encode_corpus()
queries = ['query_1', 'query_2']
passages = ["样例段落-1", "样例段落-2"]
q_embeddings = model.encode_queries(queries)
p_embeddings = model.encode(passages)
scores = q_embeddings @ p_embeddings.T
The value of argument
query_instruction_for_retrieval
see
Model List
.
FlagModel will use all available GPUs when encoding, please set
os.environ["CUDA_VISIBLE_DEVICES"]
to choose GPU.
Using this model also is easy when you have sentence-transformers installed:
pip install -U sentence-transformers
from sentence_transformers import SentenceTransformer
sentences = ["样例数据-1", "样例数据-2"]
model = SentenceTransformer('Supabase/bge-small-en')
embeddings = model.encode(sentences, normalize_embeddings=True)
print(embeddings)
For retrieval task, each query should start with an instruction (instructions see Model List ).
from sentence_transformers import SentenceTransformer
queries = ["手机开不了机怎么办?"]
passages = ["样例段落-1", "样例段落-2"]
instruction = "为这个句子生成表示以用于检索相关文章:"
model = SentenceTransformer('Supabase/bge-small-en')
q_embeddings = model.encode([instruction+q for q in queries], normalize_embeddings=True)
p_embeddings = model.encode(passages, normalize_embeddings=True)
scores = q_embeddings @ p_embeddings.T
With transformers package, you can use the model like this: First, you pass your input through the transformer model, then you select the last hidden state of first token (i.e., [CLS]) as the sentence embedding.
from transformers import AutoTokenizer, AutoModel
import torch
# Sentences we want sentence embeddings for
sentences = ["样例数据-1", "样例数据-2"]
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('Supabase/bge-small-en')
model = AutoModel.from_pretrained('Supabase/bge-small-en')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# for retrieval task, add an instruction to query
# encoded_input = tokenizer([instruction + q for q in queries], padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, cls pooling.
sentence_embeddings = model_output[0][:, 0]
# normalize embeddings
sentence_embeddings = torch.nn.functional.normalize(sentence_embeddings, p=2, dim=1)
print("Sentence embeddings:", sentence_embeddings)
This model can be used with JavaScript via Transformers.js .
import { serve } from 'https://deno.land/std@0.168.0/http/server.ts'
import { env, pipeline } from 'https://cdn.jsdelivr.net/npm/@xenova/transformers@2.5.0'
// Configuration for Deno runtime
env.useBrowserCache = false;
env.allowLocalModels = false;
const pipe = await pipeline(
'feature-extraction',
'Supabase/bge-small-en',
);
serve(async (req) => {
// Extract input string from JSON body
const { input } = await req.json();
// Generate the embedding from the user input
const output = await pipe(input, {
pooling: 'mean',
normalize: true,
});
// Extract the embedding output
const embedding = Array.from(output.data);
// Return the embedding
return new Response(
JSON.stringify({ embedding }),
{ headers: { 'Content-Type': 'application/json' } }
);
});
<script type="module">
import { pipeline } from 'https://cdn.jsdelivr.net/npm/@xenova/transformers@2.5.0';
const pipe = await pipeline(
'feature-extraction',
'Supabase/bge-small-en',
);
// Generate the embedding from text
const output = await pipe('Hello world', {
pooling: 'mean',
normalize: true,
});
// Extract the embedding output
const embedding = Array.from(output.data);
console.log(embedding);
</script>
import { pipeline } from '@xenova/transformers';
const pipe = await pipeline(
'feature-extraction',
'Supabase/bge-small-en',
);
// Generate the embedding from text
const output = await pipe('Hello world', {
pooling: 'mean',
normalize: true,
});
// Extract the embedding output
const embedding = Array.from(output.data);
console.log(embedding);
baai-general-embedding
models achieve
state-of-the-art performance on both MTEB and C-MTEB leaderboard!
More details and evaluation tools see our
scripts
.
Model Name | Dimension | Sequence Length | Average (56) | Retrieval (15) | Clustering (11) | Pair Classification (3) | Reranking (4) | STS (10) | Summarization (1) | Classification (12) |
---|---|---|---|---|---|---|---|---|---|---|
bge-large-en | 1024 | 512 | 63.98 | 53.9 | 46.98 | 85.8 | 59.48 | 81.56 | 32.06 | 76.21 |
bge-base-en | 768 | 512 | 63.36 | 53.0 | 46.32 | 85.86 | 58.7 | 81.84 | 29.27 | 75.27 |
gte-large | 1024 | 512 | 63.13 | 52.22 | 46.84 | 85.00 | 59.13 | 83.35 | 31.66 | 73.33 |
gte-base | 768 | 512 | 62.39 | 51.14 | 46.2 | 84.57 | 58.61 | 82.3 | 31.17 | 73.01 |
e5-large-v2 | 1024 | 512 | 62.25 | 50.56 | 44.49 | 86.03 | 56.61 | 82.05 | 30.19 | 75.24 |
bge-small-en | 384 | 512 | 62.11 | 51.82 | 44.31 | 83.78 | 57.97 | 80.72 | 30.53 | 74.37 |
instructor-xl | 768 | 512 | 61.79 | 49.26 | 44.74 | 86.62 | 57.29 | 83.06 | 32.32 | 61.79 |
e5-base-v2 | 768 | 512 | 61.5 | 50.29 | 43.80 | 85.73 | 55.91 | 81.05 | 30.28 | 73.84 |
bge-small-en | 384 | 512 | 61.36 | 49.46 | 44.89 | 83.54 | 57.7 | 82.07 | 30.42 | 72.31 |
text-embedding-ada-002 | 1536 | 8192 | 60.99 | 49.25 | 45.9 | 84.89 | 56.32 | 80.97 | 30.8 | 70.93 |
e5-small-v2 | 384 | 512 | 59.93 | 49.04 | 39.92 | 84.67 | 54.32 | 80.39 | 31.16 | 72.94 |
sentence-t5-xxl | 768 | 512 | 59.51 | 42.24 | 43.72 | 85.06 | 56.42 | 82.63 | 30.08 | 73.42 |
all-mpnet-base-v2 | 768 | 514 | 57.78 | 43.81 | 43.69 | 83.04 | 59.36 | 80.28 | 27.49 | 65.07 |
sgpt-bloom-7b1-msmarco | 4096 | 2048 | 57.59 | 48.22 | 38.93 | 81.9 | 55.65 | 77.74 | 33.6 | 66.19 |
all-MiniLM-L12-v2 | 384 | 512 | 56.53 | 42.69 | 41.81 | 82.41 | 58.44 | 79.8 | 27.9 | 63.21 |
all-MiniLM-L6-v2 | 384 | 512 | 56.26 | 41.95 | 42.35 | 82.37 | 58.04 | 78.9 | 30.81 | 63.05 |
contriever-base-msmarco | 768 | 512 | 56.00 | 41.88 | 41.1 | 82.54 | 53.14 | 76.51 | 30.36 | 66.68 |
sentence-t5-base | 768 | 512 | 55.27 | 33.63 | 40.21 | 85.18 | 53.09 | 81.14 | 31.39 | 69.81 |
Model | Embedding dimension | Avg | Retrieval | STS | PairClassification | Classification | Reranking | Clustering |
---|---|---|---|---|---|---|---|---|
bge-large-zh | 1024 | 64.20 | 71.53 | 53.23 | 78.94 | 72.26 | 65.11 | 48.39 |
bge-large-zh-noinstruct | 1024 | 63.53 | 70.55 | 50.98 | 76.77 | 72.49 | 64.91 | 50.01 |
BAAI/bge-base-zh | 768 | 62.96 | 69.53 | 52.05 | 77.5 | 70.98 | 64.91 | 47.63 |
BAAI/bge-small-zh | 512 | 58.27 | 63.07 | 46.87 | 70.35 | 67.78 | 61.48 | 45.09 |
m3e-base | 768 | 57.10 | 56.91 | 48.15 | 63.99 | 70.28 | 59.34 | 47.68 |
m3e-large | 1024 | 57.05 | 54.75 | 48.64 | 64.3 | 71.22 | 59.66 | 48.88 |
text-embedding-ada-002(OpenAI) | 1536 | 53.02 | 52.0 | 40.61 | 69.56 | 67.38 | 54.28 | 45.68 |
luotuo | 1024 | 49.37 | 44.4 | 39.41 | 66.62 | 65.29 | 49.25 | 44.39 |
text2vec | 768 | 47.63 | 38.79 | 41.71 | 67.41 | 65.18 | 49.45 | 37.66 |
text2vec-large | 1024 | 47.36 | 41.94 | 41.98 | 70.86 | 63.42 | 49.16 | 30.02 |
This section will introduce the way we used to train the general embedding. The training scripts are in FlagEmbedding , and we provide some examples to do pre-train and fine-tune .
1. RetroMAE Pre-train
We pre-train the model following the method
retromae
,
which shows promising improvement in retrieval task (
paper
).
The pre-training was conducted on 24 A100(40G) GPUs with a batch size of 720.
In retromae, the mask ratio of encoder and decoder are 0.3, and 0.5 respectively.
We used the AdamW optimizer and the learning rate is 2e-5.
Pre-training data :
2. Finetune
We fine-tune the model using a contrastive objective.
The format of input data is a triple
(query, positive, negative)
.
Besides the negative in the triple, we also adopt in-batch negatives strategy.
We employ the cross-device negatives sharing method to share negatives among different GPUs,
which can dramatically
increase the number of negatives
.
We trained our model on 48 A100(40G) GPUs with a large batch size of 32,768 (so there are 65,535 negatives for each query in a batch). We used the AdamW optimizer and the learning rate is 1e-5. The temperature for contrastive loss is 0.01.
For the version with
*-instrcution
, we add instruction to the query for retrieval task in the training.
For english, the instruction is
Represent this sentence for searching relevant passages:
;
For chinese, the instruction is
为这个句子生成表示以用于检索相关文章:
.
In the evaluation, the instruction should be added for sentence to passages retrieval task, not be added for other tasks.
The finetune script is accessible in this repository: FlagEmbedding . You can easily finetune your model with it.
Training data :
For English, we collect 230M text pairs from wikipedia , cc-net , and so on.
For chinese, we collect 120M text pairs from wudao , simclue and so on.
The data collection is to be released in the future.
We will continually update the embedding models and training codes, hoping to promote the development of the embedding model community.
FlagEmbedding is licensed under MIT License . The released models can be used for commercial purposes free of charge.
bge-small-en huggingface.co is an AI model on huggingface.co that provides bge-small-en's model effect (), which can be used instantly with this Supabase bge-small-en model. huggingface.co supports a free trial of the bge-small-en model, and also provides paid use of the bge-small-en. Support call bge-small-en model through api, including Node.js, Python, http.
bge-small-en huggingface.co is an online trial and call api platform, which integrates bge-small-en's modeling effects, including api services, and provides a free online trial of bge-small-en, you can try bge-small-en online for free by clicking the link below.
bge-small-en is an open source model from GitHub that offers a free installation service, and any user can find bge-small-en on GitHub to install. At the same time, huggingface.co provides the effect of bge-small-en install, users can directly use bge-small-en installed effect in huggingface.co for debugging and trial. It also supports api for free installation.