writinwaters
commited on
Commit
·
ddeb877
1
Parent(s):
75a07ce
Expanded list of locally deployed embedding models (#930)
Browse files### What problem does this PR solve?
_Briefly describe what this PR aims to solve. Include background context
that will help reviewers understand the purpose of the PR._
### Type of change
- [x] Documentation Update
- README.md +4 -0
- README_ja.md +4 -0
- README_zh.md +4 -0
- docs/guides/configure_knowledge_base.md +1 -0
- docs/references/faq.md +13 -0
README.md
CHANGED
@@ -28,6 +28,10 @@
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. It offers a streamlined RAG workflow for businesses of any scale, combining LLM (Large Language Models) to provide truthful question-answering capabilities, backed by well-founded citations from various complex formatted data.
|
30 |
|
|
|
|
|
|
|
|
|
31 |
## 📌 Latest Updates
|
32 |
|
33 |
- 2024-05-23 Supports [RAPTOR](https://arxiv.org/html/2401.18059v1) for better text retrieval.
|
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. It offers a streamlined RAG workflow for businesses of any scale, combining LLM (Large Language Models) to provide truthful question-answering capabilities, backed by well-founded citations from various complex formatted data.
|
30 |
|
31 |
+
## 🎮 Demo
|
32 |
+
|
33 |
+
Visit our demo at [https://demo.ragflow.io](https://demo.ragflow.io)
|
34 |
+
|
35 |
## 📌 Latest Updates
|
36 |
|
37 |
- 2024-05-23 Supports [RAPTOR](https://arxiv.org/html/2401.18059v1) for better text retrieval.
|
README_ja.md
CHANGED
@@ -28,6 +28,10 @@
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) は、深い文書理解に基づいたオープンソースの RAG (Retrieval-Augmented Generation) エンジンである。LLM(大規模言語モデル)を組み合わせることで、様々な複雑なフォーマットのデータから根拠のある引用に裏打ちされた、信頼できる質問応答機能を実現し、あらゆる規模のビジネスに適した RAG ワークフローを提供します。
|
30 |
|
|
|
|
|
|
|
|
|
31 |
## 📌 最新情報
|
32 |
|
33 |
- 2024-05-23 より良いテキスト検索のために[RAPTOR](https://arxiv.org/html/2401.18059v1)をサポート。
|
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) は、深い文書理解に基づいたオープンソースの RAG (Retrieval-Augmented Generation) エンジンである。LLM(大規模言語モデル)を組み合わせることで、様々な複雑なフォーマットのデータから根拠のある引用に裏打ちされた、信頼できる質問応答機能を実現し、あらゆる規模のビジネスに適した RAG ワークフローを提供します。
|
30 |
|
31 |
+
## 🎮 Demo
|
32 |
+
|
33 |
+
デモをお試しください:[https://demo.ragflow.io](https://demo.ragflow.io)。
|
34 |
+
|
35 |
## 📌 最新情報
|
36 |
|
37 |
- 2024-05-23 より良いテキスト検索のために[RAPTOR](https://arxiv.org/html/2401.18059v1)をサポート。
|
README_zh.md
CHANGED
@@ -28,6 +28,10 @@
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) 是一款基于深度文档理解构建的开源 RAG(Retrieval-Augmented Generation)引擎。RAGFlow 可以为各种规模的企业及个人提供一套精简的 RAG 工作流程,结合大语言模型(LLM)针对用户各类不同的复杂格式数据提供可靠的问答以及有理有据的引用。
|
30 |
|
|
|
|
|
|
|
|
|
31 |
## 📌 近期更新
|
32 |
|
33 |
- 2024-05-23 实现 [RAPTOR](https://arxiv.org/html/2401.18059v1) 提供更好的文本检索。
|
|
|
28 |
|
29 |
[RAGFlow](https://ragflow.io/) 是一款基于深度文档理解构建的开源 RAG(Retrieval-Augmented Generation)引擎。RAGFlow 可以为各种规模的企业及个人提供一套精简的 RAG 工作流程,结合大语言模型(LLM)针对用户各类不同的复杂格式数据提供可靠的问答以及有理有据的引用。
|
30 |
|
31 |
+
## 🎮 Demo 试用
|
32 |
+
|
33 |
+
请登录网址 [https://demo.ragflow.io](https://demo.ragflow.io) 试用 demo。
|
34 |
+
|
35 |
## 📌 近期更新
|
36 |
|
37 |
- 2024-05-23 实现 [RAPTOR](https://arxiv.org/html/2401.18059v1) 提供更好的文本检索。
|
docs/guides/configure_knowledge_base.md
CHANGED
@@ -62,6 +62,7 @@ An embedding model builds vector index on file chunks. Once you have chosen an e
|
|
62 |
|
63 |
The following embedding models can be deployed locally:
|
64 |
|
|
|
65 |
- BAAI/bge-base-en-v1.5
|
66 |
- BAAI/bge-large-en-v1.5
|
67 |
- BAAI/bge-small-en-v1.5
|
|
|
62 |
|
63 |
The following embedding models can be deployed locally:
|
64 |
|
65 |
+
- BAAI/bge-large-zh-v1.5
|
66 |
- BAAI/bge-base-en-v1.5
|
67 |
- BAAI/bge-large-en-v1.5
|
68 |
- BAAI/bge-small-en-v1.5
|
docs/references/faq.md
CHANGED
@@ -18,6 +18,19 @@ The "garbage in garbage out" status quo remains unchanged despite the fact that
|
|
18 |
|
19 |
English, simplified Chinese, traditional Chinese for now.
|
20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
## Performance
|
22 |
|
23 |
### 1. Why does it take longer for RAGFlow to parse a document than LangChain?
|
|
|
18 |
|
19 |
English, simplified Chinese, traditional Chinese for now.
|
20 |
|
21 |
+
### 3. Which embedding models can be deployed locally?
|
22 |
+
|
23 |
+
- BAAI/bge-large-zh-v1.5
|
24 |
+
- BAAI/bge-base-en-v1.5
|
25 |
+
- BAAI/bge-large-en-v1.5
|
26 |
+
- BAAI/bge-small-en-v1.5
|
27 |
+
- BAAI/bge-small-zh-v1.5
|
28 |
+
- jinaai/jina-embeddings-v2-base-en
|
29 |
+
- jinaai/jina-embeddings-v2-small-en
|
30 |
+
- nomic-ai/nomic-embed-text-v1.5
|
31 |
+
- sentence-transformers/all-MiniLM-L6-v2
|
32 |
+
- maidalun1020/bce-embedding-base_v1
|
33 |
+
|
34 |
## Performance
|
35 |
|
36 |
### 1. Why does it take longer for RAGFlow to parse a document than LangChain?
|