bert4torch Download - bert4torch Source code download

bert4torch

Other source code

v0.5.4

Download

bert4torch

Documentation | Torch4keras | Examples | build_MiniLLM_from_scratch | bert4vector

1. Download and install

Install stable version

pip install bert4torch

Install the latest version

pip install git+https://github.com/Tongjilibo/bert4torch

Note : The release of pip packages is slower than the development version on git. Pay attention to the reference path of git clone and pay attention to whether the weight needs to be converted.
Test case : git clone https://github.com/Tongjilibo/bert4torch , modify the pretrained model file path and data path in the example to start the script
Self-training : modify the corresponding data processing code block for your own data
Development environment : Originally used to develop with torch==1.10 version, and has now switched to torch2.0 development. If other versions encounter disagreements, please feel free to feedback.

2. Function

LLM model : Load open source big model weights such as chatglm, llama, baichuan, ziya, bloom, etc. for inference and fine-tuning, and deploy big models in one line on the command line.
Core functions : Loading pre-training weights such as bert, roberta, albert, xlnet, nezha, bart, RoFormer, RoFormer_V2, ELECTRA, GPT, GPT2, T5, GAU-alpha, ERNIE, etc. continue to finetune, and support flexibly defining your own model based on bert
Rich examples : including llm, pretrain, sentence_classification, sentence_embedding, sequence_labeling, relationship_extraction, seq2seq, serving and other solutions
Experimental verification : Experimental verification has been made on the public data set, using the following examples data set and experimental indicators
Easy-to-use trick : integrates common tricks, plug and play
Other features : load the transformers library model together; call method is simple and efficient; dynamic display of training progress bars; print parameter volume with torchinfo; default Logger and Tensorboard are easy to record the training process; custom fit process to meet high-level needs
Training process :

Function	bert4torch	transformers	Remark
Training progress bar	✅	✅	Progress bar prints loss and defined metrics
Distributed training dp/ddp	✅	✅	Torch comes with dp/ddp
Various callbacks	✅	✅	Log/tensorboard/earlystop/wandb, etc.
Big model reasoning, stream/batch output	✅	✅	Each model is universal and does not require separate maintenance of scripts
Large model fine tune	✅	✅	Lora depends on the peft library, PV2 comes with its own
Rich tricks	✅		Tricks plug and play against training and other tricks
The code is simple and easy to understand, and the custom space is large	✅		High code reuse, keras code training style
Warehouse maintenance capability/influence/usage/compatibility		✅	Currently, personal maintenance of the warehouse
One-click deployment of large models

3. Get started quickly

3.1 Get started tutorial

Quick-Start
Quick start tutorials, tutorial examples, practical examples
Introduction to bert4torch (Zhihu), get started quickly with bert4torch (Zhihu), and bert4torch has been updated again (Zhihu)

3.2 Quickly deploy large model services on the command line

Local/Internet loading

 # 联网下载全部文件
bert4torch-llm-server --checkpoint_path Qwen2-0.5B-Instruct

# 加载本地大模型，联网下载bert4torch_config.json
bert4torch-llm-server --checkpoint_path /data/pretrain_ckpt/Qwen/Qwen2-0.5B-Instruct --config_path Qwen/Qwen2-0.5B-Instruct

# 加载本地大模型，且bert4torch_config.json已经下载并放于同名目录下
bert4torch-llm-server --checkpoint_path /data/pretrain_ckpt/Qwen/Qwen2-0.5B-Instruct

Command line/gradio webpage/openai_api

 # 命令行
bert4torch-llm-server --checkpoint_path /data/pretrain_ckpt/Qwen/Qwen2-0.5B-Instruct --mode cli

# gradio网页
bert4torch-llm-server --checkpoint_path /data/pretrain_ckpt/Qwen/Qwen2-0.5B-Instruct --mode gradio

# openai_api
bert4torch-llm-server --checkpoint_path /data/pretrain_ckpt/Qwen/Qwen2-0.5B-Instruct --mode openai

Command line chat example

4. Version and update history

Version 4.1 History

Update date	bert4torch	torch4keras	Version Description
20240928	0.5.4	0.2.7	[New Features] Add deepseek series, MiniCPM, MiniCPMV, llama3.2, Qwen2.5; support device_map=auto; [Fix] Fix batch_generate and n>1 bugs
20240814	0.5.3	0.2.6	【New Features】Add llama3.1/Yi1.5; Automatically select download from hfmirror; support command line parameters `bert4torch-llm-server`
20240801	0.5.2	0.2.5	[New function] Chatglm/qwen series supports function call calls, and adds internlm2 series; [Small optimization] Simplify the call of chat demo in pipeline, generate token element is allowed to be a list, unify the rope_scaling parameter name, and add rope derived classes; [bug] Fix flash_attn2 inference bug, fix bart's tie_word_embedding bug

More versions

4.2 Update history

More history

5. Pre-training weights

Pre-trained models support multiple code loading methods

 from bert4torch . models import build_transformer_model

# 1. 仅指定config_path: 从头初始化模型结构, 不加载预训练模型
model = build_transformer_model ( './model/bert4torch_config.json' )

# 2. 仅指定checkpoint_path: 
## 2.1 文件夹路径: 自动寻找路径下的*.bin/*.safetensors权重文件 + 需把bert4torch_config.json下载并放于该目录下
model = build_transformer_model ( checkpoint_path = './model' )

## 2.2 文件路径/列表: 文件路径即权重路径/列表, bert4torch_config.json会从同级目录下寻找
model = build_transformer_model ( checkpoint_path = './pytorch_model.bin' )

## 2.3 model_name: hf上预训练权重名称, 会自动下载hf权重以及bert4torch_config.json文件
model = build_transformer_model ( checkpoint_path = 'bert-base-chinese' )

# 3. 同时指定config_path和checkpoint_path(本地路径名或model_name排列组合): 
#    本地路径从本地加载，pretrained_model_name会联网下载
config_path = './model/bert4torch_config.json'  # 或'bert-base-chinese'
checkpoint_path = './model/pytorch_model.bin'  # 或'bert-base-chinese'
model = build_transformer_model ( config_path , checkpoint_path )

Pretrained weight link and bert4torch_config.json

*Note:

The representation of高亮格式(such as bert-base-chinese ) can be directly build_transformer_model() to download online
Domestic mirror websites are accelerated to download
- HF_ENDPOINT=https://hf-mirror.com python your_script.py
- export HF_ENDPOINT=https://hf-mirror.com before executing python code
- Set as follows at the beginning of the python code
```
 import os
os . environ [ 'HF_ENDPOINT' ] = "https://hf-mirror.com" 
```

6. Thanks

Thanks to Su Shen for the bet4keras implemented. There are many places in this implementation that refer to the source code of bet4keras. I would like to sincerely thank the boss for his selfless dedication;
Secondly, thanks to the project bet4pytorch, and under the guidance of this project, I gave me the ideas and ideas of using pytorch to reproduce bet4keras.

7. Quote

 @misc{bert4torch,
  title={bert4torch},
  author={Bo Li},
  year={2022},
  howpublished={url{https://github.com/Tongjilibo/bert4torch}},
}

8. Others

Wechat & Star History Chart
The number of WeChat groups exceeds 200 (with invitation restrictions), you can add personal WeChat groups

WeChat ID

WeChat group

Star History Chart

Expand

Additional Information

Version v0.5.4
Type Other source code
Update Time 2025-04-19
size 3.42MB
From Github

Related Applications

Google Dorks

2025-03-10
shepherd

2025-06-04
mongo express

2025-06-04
hidusbf

2025-02-14
Free Algorithms Books

2025-05-29
markdownpedia

2025-04-22

Recommended for You

chat.petals.dev

Other source code

1.0.0
GPT Prompt Templates

Other source code

1.0.0
GPTyped

Other source code

GPTyped 1.0.5
Google Dorks

Other source code

1.0
shepherd

Other source code

v6.1.6-react-shepherd: Prepare Release (#3063)
mongo express

Other source code

v1.1.0-rc-3
Google Dorks

Other source code

1.0
shepherd

Other source code

v6.1.6-react-shepherd: Prepare Release (#3063)
mongo express

Other source code

v1.1.0-rc-3

Related Information All

Model classification	Model name	Source of weight	Weight link/checkpoint_path	config_path
bert	bert-base-chinese	google-bert	`bert-base-chinese`	`bert-base-chinese`
	chinese_L-12_H-768_A-12	Google	TF weight `Tongjilibo/bert-chinese_L-12_H-768_A-12`
	chinese-bert-wwm-ext	HFL	`hfl/chinese-bert-wwm-ext`	`hfl/chinese-bert-wwm-ext`
	bert-base-multilingual-cased	google-bert	`bert-base-multilingual-cased`	`bert-base-multilingual-cased`
	MacBERT	HFL	`hfl/chinese-macbert-base` `hfl/chinese-macbert-large`	`hfl/chinese-macbert-base` `hfl/chinese-macbert-large`
	WoBERT	Zhuyi Technology	`junnyu/wobert_chinese_base` , `junnyu/wobert_chinese_plus_base`	`junnyu/wobert_chinese_base` `junnyu/wobert_chinese_plus_base`
roberta	chinese-roberta-wwm-ext	HFL	`hfl/chinese-roberta-wwm-ext` `hfl/chinese-roberta-wwm-ext-large` (Large's mlm weight is randomly initialized)	`hfl/chinese-roberta-wwm-ext` `hfl/chinese-roberta-wwm-ext-large`
	roberta-small/tiny	Zhuyi Technology	`Tongjilibo/chinese_roberta_L-4_H-312_A-12` `Tongjilibo/chinese_roberta_L-6_H-384_A-12`
	roberta-base	FacebookAI	`roberta-base`	`roberta-base`
	guwenbert	ethanyt	`ethanyt/guwenbert-base`	`ethanyt/guwenbert-base`
albert	albert_zh albert_pytorch	brightmart	`voidful/albert_chinese_tiny` `voidful/albert_chinese_small` `voidful/albert_chinese_base` `voidful/albert_chinese_large` `voidful/albert_chinese_xlarge` `voidful/albert_chinese_xxlarge`	`voidful/albert_chinese_tiny` `voidful/albert_chinese_small` `voidful/albert_chinese_base` `voidful/albert_chinese_large` `voidful/albert_chinese_xlarge` `voidful/albert_chinese_xxlarge`
nezha	NEZHA NeZha_Chinese_PyTorch	huawei_noah	`sijunhe/nezha-cn-base` `sijunhe/nezha-cn-large` `sijunhe/nezha-base-wwm` `sijunhe/nezha-large-wwm`	`sijunhe/nezha-cn-base` `sijunhe/nezha-cn-large` `sijunhe/nezha-base-wwm` `sijunhe/nezha-large-wwm`
	nezha_gpt_dialog	bojone	`Tongjilibo/nezha_gpt_dialog`
xlnet	Chinese-XLNet	HFL	`hfl/chinese-xlnet-base`	`hfl/chinese-xlnet-base`
	transformer_xl	huggingface	`transfo-xl/transfo-xl-wt103`	`transfo-xl/transfo-xl-wt103`
deberta	Erlangshen-DeBERTa-v2	IDEA	`IDEA-CCNL/Erlangshen-DeBERTa-v2-97M-Chinese` `IDEA-CCNL/Erlangshen-DeBERTa-v2-320M-Chinese` `IDEA-CCNL/Erlangshen-DeBERTa-v2-710M-Chinese`	`IDEA-CCNL/Erlangshen-DeBERTa-v2-97M-Chinese` `IDEA-CCNL/Erlangshen-DeBERTa-v2-320M-Chinese` `IDEA-CCNL/Erlangshen-DeBERTa-v2-710M-Chinese`
electra	Chinese-ELECTRA	HFL	`hfl/chinese-electra-base-discriminator`	`hfl/chinese-electra-base-discriminator`
ernie	ernie	Baidu Wenxin	`nghuyong/ernie-1.0-base-zh` `nghuyong/ernie-3.0-base-zh`	`nghuyong/ernie-1.0-base-zh` `nghuyong/ernie-3.0-base-zh`
roformer	roformer	Zhuyi Technology	`junnyu/roformer_chinese_base`	`junnyu/roformer_chinese_base`
	roformer_v2	Zhuyi Technology	`junnyu/roformer_v2_chinese_char_base`	`junnyu/roformer_v2_chinese_char_base`
simbert	simbert	Zhuyi Technology	`Tongjilibo/simbert-chinese-base` `Tongjilibo/simbert-chinese-small` `Tongjilibo/simbert-chinese-tiny`
	simbert_v2/roformer-sim	Zhuyi Technology	`junnyu/roformer_chinese_sim_char_base` , `junnyu/roformer_chinese_sim_char_ft_base` , `junnyu/roformer_chinese_sim_char_small` , `junnyu/roformer_chinese_sim_char_ft_small`	`junnyu/roformer_chinese_sim_char_base` `junnyu/roformer_chinese_sim_char_ft_base` `junnyu/roformer_chinese_sim_char_small` `junnyu/roformer_chinese_sim_char_ft_small`
gau	GAU-alpha	Zhuyi Technology	`Tongjilibo/chinese_GAU-alpha-char_L-24_H-768`
Uie	Uie uie_pytorch	Baidu	`Tongjilibo/uie-base`
gpt	CDial-GPT	thu-coai	`thu-coai/CDial-GPT_LCCC-base` `thu-coai/CDial-GPT_LCCC-large`	`thu-coai/CDial-GPT_LCCC-base` `thu-coai/CDial-GPT_LCCC-large`
	cmp_lm(2.6 billion)	Tsinghua	`TsinghuaAI/CPM-Generate`	`TsinghuaAI/CPM-Generate`
	nezha_gen	huawei_noah	`Tongjilibo/chinese_nezha_gpt_L-12_H-768_A-12`
	gpt2-chinese-cluecorpussmall	UER	`uer/gpt2-chinese-cluecorpussmall`	`uer/gpt2-chinese-cluecorpussmall`
	gpt2-ml	imcaspar	torrh BaiduYun(84dh)	`gpt2-ml_15g_corpus` `gpt2-ml_30g_corpus`
bart	bart_base_chinese	Fudan fnlp	`fnlp/bart-base-chinese` v1.0	`fnlp/bart-base-chinese` `fnlp/bart-base-chinese-v1.0`
t5	t5	UER	`uer/t5-small-chinese-cluecorpussmall` `uer/t5-base-chinese-cluecorpussmall`	`uer/t5-base-chinese-cluecorpussmall` `uer/t5-small-chinese-cluecorpussmall`
	mt5	Google	`google/mt5-base`	`google/mt5-base`
	t5_pegasus	Zhuyi Technology	`Tongjilibo/chinese_t5_pegasus_small` `Tongjilibo/chinese_t5_pegasus_base`
	chatyuan	clue-ai	`ClueAI/ChatYuan-large-v1` `ClueAI/ChatYuan-large-v2`	`ClueAI/ChatYuan-large-v1` `ClueAI/ChatYuan-large-v2`
	PromptCLUE	clue-ai	`ClueAI/PromptCLUE-base`	`ClueAI/PromptCLUE-base`
chatglm	chatglm-6b	THUDM	`THUDM/chatglm-6b` `THUDM/chatglm-6b-int8` `THUDM/chatglm-6b-int4` v0.1.0	`THUDM/chatglm-6b` `THUDM/chatglm-6b-int8` `THUDM/chatglm-6b-int4` `THUDM/chatglm-6b-v0.1.0`
	chatglm2-6b	THUDM	`THUDM/chatglm2-6b` `THUDM/chatglm2-6b-int4` `THUDM/chatglm2-6b-32k`	`THUDM/chatglm2-6b` `THUDM/chatglm2-6b-int4` `THUDM/chatglm2-6b-32k`
	chatglm3-6b	THUDM	`THUDM/chatglm3-6b` `THUDM/chatglm3-6b-32k`	`THUDM/chatglm3-6b` `THUDM/chatglm3-6b-32k`
	glm4-9b	THUDM	`THUDM/glm-4-9b` `THUDM/glm-4-9b-chat` `THUDM/glm-4-9b-chat-1m`	`THUDM/glm-4-9b` `THUDM/glm-4-9b-chat` `THUDM/glm-4-9b-chat-1m`
llama	llama	meta		`meta-llama/llama-7b` `meta-llama/llama-13b`
	llama-2	meta	meta-llama/Llama-2-7b-hf meta-llama/Llama-2-7b-chat-hf meta-llama/Llama-2-13b-hf meta-llama/Llama-2-13b-chat-hf	`meta-llama/Llama-2-7b-hf` `meta-llama/Llama-2-7b-chat-hf` `meta-llama/Llama-2-13b-hf` `meta-llama/Llama-2-13b-chat-hf`
	llama-3	meta	`meta-llama/Meta-Llama-3-8B` `meta-llama/Meta-Llama-3-8B-Instruct`	`meta-llama/Meta-Llama-3-8B` `meta-llama/Meta-Llama-3-8B-Instruct`
	llama-3.1	meta	`meta-llama/Meta-Llama-3.1-8B` `meta-llama/Meta-Llama-3.1-8B-Instruct`	`meta-llama/Meta-Llama-3.1-8B` `meta-llama/Meta-Llama-3.1-8B-Instruct`
	llama-3.2	meta	`meta-llama/Llama-3.2-1B` `meta-llama/Llama-3.2-1B-Instruct` `meta-llama/Llama-3.2-3B` `meta-llama/Llama-3.2-3B-Instruct`	`meta-llama/Llama-3.2-1B` `meta-llama/Llama-3.2-1B-Instruct` `meta-llama/Llama-3.2-3B` `meta-llama/Llama-3.2-3B-Instruct`
	Chinese-LLaMA-Alpaca	HFL		`hfl/chinese_alpaca_plus_7b` `hfl/chinese_llama_plus_7b`
	Chinese-LLaMA-Alpaca-2	HFL		To be added
	Chinese-LLaMA-Alpaca-3	HFL		To be added
	Belle_llama	LianjiaTech	BelleGroup/BELLE-LLaMA-7B-2M-enc	Synthesis instructions, `BelleGroup/BELLE-LLaMA-7B-2M-enc`
	Ziya	IDEA-CCNL	IDEA-CCNL/Ziya-LLaMA-13B-v1 IDEA-CCNL/Ziya-LLaMA-13B-v1.1 IDEA-CCNL/Ziya-LLaMA-13B-Pretrain-v1	`IDEA-CCNL/Ziya-LLaMA-13B-v1` `IDEA-CCNL/Ziya-LLaMA-13B-v1.1`
	vicuna	lmsys	`lmsys/vicuna-7b-v1.5`	`lmsys/vicuna-7b-v1.5`
Baichuan	Baichuan	baichuan-inc	`baichuan-inc/Baichuan-7B` `baichuan-inc/Baichuan-13B-Base` `baichuan-inc/Baichuan-13B-Chat`	`baichuan-inc/Baichuan-7B` `baichuan-inc/Baichuan-13B-Base` `baichuan-inc/Baichuan-13B-Chat`
	Baichuan2	baichuan-inc	`baichuan-inc/Baichuan2-7B-Base` `baichuan-inc/Baichuan2-7B-Chat` `baichuan-inc/Baichuan2-13B-Base` `baichuan-inc/Baichuan2-13B-Chat`	`baichuan-inc/Baichuan2-7B-Base` `baichuan-inc/Baichuan2-7B-Chat` `baichuan-inc/Baichuan2-13B-Base` `baichuan-inc/Baichuan2-13B-Chat`
Yi	Yi	01-ai	`01-ai/Yi-6B` `01-ai/Yi-6B-200K` `01-ai/Yi-9B` `01-ai/Yi-9B-200K`	`01-ai/Yi-6B` `01-ai/Yi-6B-200K` `01-ai/Yi-9B` `01-ai/Yi-9B-200K`
	Yi-1.5	01-ai	`01-ai/Yi-1.5-6B` `01-ai/Yi-1.5-6B-Chat` `01-ai/Yi-1.5-9B` `01-ai/Yi-1.5-9B-32K` `01-ai/Yi-1.5-9B-Chat` `01-ai/Yi-1.5-9B-Chat-16K`	`01-ai/Yi-1.5-6B` `01-ai/Yi-1.5-6B-Chat` `01-ai/Yi-1.5-9B` `01-ai/Yi-1.5-9B-32K` `01-ai/Yi-1.5-9B-Chat` `01-ai/Yi-1.5-9B-Chat-16K`
bloom	bloom	bigscience	`bigscience/bloom-560m` `bigscience/bloomz-560m`	`bigscience/bloom-560m` `bigscience/bloomz-560m`
Qwen	Qwen	Alibaba Cloud	`Qwen/Qwen-1_8B` `Qwen/Qwen-1_8B-Chat` `Qwen/Qwen-7B` `Qwen/Qwen-7B-Chat` `Qwen/Qwen-14B` `Qwen/Qwen-14B-Chat`	`Qwen/Qwen-1_8B` `Qwen/Qwen-1_8B-Chat` `Qwen/Qwen-7B` `Qwen/Qwen-7B-Chat` `Qwen/Qwen-14B` `Qwen/Qwen-14B-Chat`
	Qwen1.5	Alibaba Cloud	`Qwen/Qwen1.5-0.5B` `Qwen/Qwen1.5-0.5B-Chat` `Qwen/Qwen1.5-1.8B` `Qwen/Qwen1.5-1.8B-Chat` `Qwen/Qwen1.5-7B` `Qwen/Qwen1.5-7B-Chat` `Qwen/Qwen1.5-14B` `Qwen/Qwen1.5-14B-Chat`	`Qwen/Qwen1.5-0.5B` `Qwen/Qwen1.5-0.5B-Chat` `Qwen/Qwen1.5-1.8B` `Qwen/Qwen1.5-1.8B-Chat` `Qwen/Qwen1.5-7B` `Qwen/Qwen1.5-7B-Chat` `Qwen/Qwen1.5-14B` `Qwen/Qwen1.5-14B-Chat`
	Qwen2	Alibaba Cloud	`Qwen/Qwen2-0.5B` `Qwen/Qwen2-0.5B-Instruct` `Qwen/Qwen2-1.5B` `Qwen/Qwen2-1.5B-Instruct` `Qwen/Qwen2-7B` `Qwen/Qwen2-7B-Instruct`	`Qwen/Qwen2-0.5B` `Qwen/Qwen2-0.5B-Instruct` `Qwen/Qwen2-1.5B` `Qwen/Qwen2-1.5B-Instruct` `Qwen/Qwen2-7B` `Qwen/Qwen2-7B-Instruct`
	Qwen2-VL	Alibaba Cloud	`Qwen/Qwen2-VL-2B-Instruct` `Qwen/Qwen2-VL-7B-Instruct`	`Qwen/Qwen2-VL-2B-Instruct` `Qwen/Qwen2-VL-7B-Instruct`
	Qwen2.5	Alibaba Cloud	`Qwen/Qwen2.5-0.5B` `Qwen/Qwen2.5-0.5B-Instruct` `Qwen/Qwen2.5-1.5B` `Qwen/Qwen2.5-1.5B-Instruct` `Qwen/Qwen2.5-3B` `Qwen/Qwen2.5-3B-Instruct` `Qwen/Qwen2.5-7B` `Qwen/Qwen2.5-7B-Instruct` `Qwen/Qwen2.5-14B` `Qwen/Qwen2.5-14B-Instruct`	`Qwen/Qwen2.5-0.5B` `Qwen/Qwen2.5-0.5B-Instruct` `Qwen/Qwen2.5-1.5B` `Qwen/Qwen2.5-1.5B-Instruct` `Qwen/Qwen2.5-3B` `Qwen/Qwen2.5-3B-Instruct` `Qwen/Qwen2.5-7B` `Qwen/Qwen2.5-7B-Instruct` `Qwen/Qwen2.5-14B` `Qwen/Qwen2.5-14B-Instruct`
InternLM	InternLM	Shanghai Artificial Intelligence Laboratory	`internlm/internlm-7b` `internlm/internlm-chat-7b`	`internlm/internlm-7b` `internlm/internlm-chat-7b`
	InternLM2	Shanghai Artificial Intelligence Laboratory	`internlm/internlm2-1_8b` `internlm/internlm2-chat-1_8b` `internlm/internlm2-7b` `internlm/internlm2-chat-7b` `internlm/internlm2-20b` `internlm/internlm2-chat-20b`	`internlm/internlm2-1_8b` `internlm/internlm2-chat-1_8b` `internlm/internlm2-7b` `internlm/internlm2-chat-7b`
	InternLM2.5	Shanghai Artificial Intelligence Laboratory	`internlm/internlm2_5-7b` `internlm/internlm2_5-7b-chat` `internlm/internlm2_5-7b-chat-1m`	`internlm/internlm2_5-7b` `internlm/internlm2_5-7b-chat` `internlm/internlm2_5-7b-chat-1m`
Falcon	Falcon	tiiuae	`tiiuae/falcon-rw-1b` `tiiuae/falcon-7b` `tiiuae/falcon-7b-instruct`	`tiiuae/falcon-rw-1b` `tiiuae/falcon-7b` `tiiuae/falcon-7b-instruct`
DeepSeek	DeepSeek-MoE	In-depth search	`deepseek-ai/deepseek-moe-16b-base` `deepseek-ai/deepseek-moe-16b-chat`	`deepseek-ai/deepseek-moe-16b-base` `deepseek-ai/deepseek-moe-16b-chat`
	DeepSeek-LLM	In-depth search	`deepseek-ai/deepseek-llm-7b-base` `deepseek-ai/deepseek-llm-7b-chat`	`deepseek-ai/deepseek-llm-7b-base` `deepseek-ai/deepseek-llm-7b-chat`
	DeepSeek-V2	In-depth search	`deepseek-ai/DeepSeek-V2-Lite` `deepseek-ai/DeepSeek-V2-Lite-Chat`	`deepseek-ai/DeepSeek-V2-Lite` `deepseek-ai/DeepSeek-V2-Lite-Chat`
	DeepSeek-Coder	In-depth search	`deepseek-ai/deepseek-coder-1.3b-base` `deepseek-ai/deepseek-coder-1.3b-instruct` `deepseek-ai/deepseek-coder-6.7b-base` `deepseek-ai/deepseek-coder-6.7b-instruct` `deepseek-ai/deepseek-coder-7b-base-v1.5` `deepseek-ai/deepseek-coder-7b-instruct-v1.5`	`deepseek-ai/deepseek-coder-1.3b-base` `deepseek-ai/deepseek-coder-1.3b-instruct` `deepseek-ai/deepseek-coder-6.7b-base` `deepseek-ai/deepseek-coder-6.7b-instruct` `deepseek-ai/deepseek-coder-7b-base-v1.5` `deepseek-ai/deepseek-coder-7b-instruct-v1.5`
	DeepSeek-Coder-V2	In-depth search	`deepseek-ai/DeepSeek-Coder-V2-Lite-Base` `deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct`	`deepseek-ai/DeepSeek-Coder-V2-Lite-Base` `deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct`
	DeepSeek-Math	In-depth search	`deepseek-ai/deepseek-math-7b-base` `deepseek-ai/deepseek-math-7b-instruct` `deepseek-ai/deepseek-math-7b-rl`	`deepseek-ai/deepseek-math-7b-base` `deepseek-ai/deepseek-math-7b-instruct` `deepseek-ai/deepseek-math-7b-rl`
MiniCPM	MiniCPM	OpenBMB	`openbmb/MiniCPM-2B-sft-bf16` `openbmb/MiniCPM-2B-dpo-bf16` `openbmb/MiniCPM-2B-128k` `openbmb/MiniCPM-1B-sft-bf16`	`openbmb/MiniCPM-2B-sft-bf16` `openbmb/MiniCPM-2B-dpo-bf16` `openbmb/MiniCPM-2B-128k` `openbmb/MiniCPM-1B-sft-bf16`
	MiniCPM-V	OpenBMB	`openbmb/MiniCPM-V-2_6` `openbmb/MiniCPM-Llama3-V-2_5`	`openbmb/MiniCPM-V-2_6` `openbmb/MiniCPM-Llama3-V-2_5`
Embedding	text2vec-base-chinese	shibing624	`shibing624/text2vec-base-chinese`	`shibing624/text2vec-base-chinese`
	m3e	moka-ai	`moka-ai/m3e-base`	`moka-ai/m3e-base`
	bge	BAAI	`BAAI/bge-large-en-v1.5` `BAAI/bge-large-zh-v1.5` `BAAI/bge-base-en-v1.5` `BAAI/bge-base-zh-v1.5` `BAAI/bge-small-en-v1.5` `BAAI/bge-small-zh-v1.5`	`BAAI/bge-large-en-v1.5` `BAAI/bge-large-zh-v1.5` `BAAI/bge-base-en-v1.5` `BAAI/bge-base-zh-v1.5` `BAAI/bge-small-en-v1.5` `BAAI/bge-small-zh-v1.5`
	gte	Thenlper	`thenlper/gte-large-zh` `thenlper/gte-base-zh`	`thenlper/gte-base-zh` `thenlper/gte-large-zh`