gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,451 Commits 2 Branches 0 Tags
19ee3d109c10d45e9501d7f3232db527e02a8c4b
Commit Graph

822 Commits

Author SHA1 Message Date
yangdx
2c8885792c Refactor /api/generate：use llm_model_func directly 2025-01-24 18:39:43 +08:00
yangdx
b94cae9990 draft implementation of /api/generate endpoint 2025-01-24 18:18:14 +08:00
hyb
7b02fac7b9 feat: New ini file reading database configuration method for production environment convenience, modified Lightrag ainsert method - add.doc_keys retrieval method to only filter existing ones, but this will prevent failed documents from being stored again. Added -- chunk_Size and -- chunkoverlapped size for production environment convenience, and added llm_binding: openai-ollama for easy use of openai while also using ollama embedding 2025-01-24 17:10:19 +08:00
ruirui
d5a6d08877 fix completion_delimiter with record_delimiter 2025-01-24 16:34:24 +08:00
zrguo
3d93df4049 Merge pull request #634 from magicyuan876/main 
修复当出现异常时，会导致更新数据卡死的bug
 2025-01-24 16:30:50 +08:00
zrguo
afb30fe6c1 Merge branch 'main' into redis_impl 2025-01-24 16:27:40 +08:00
Magic_yuan
f6d29e1793 修复缓存处理的运行时错误 
].func  # ["func"]
 ^^^^
AttributeError: 'dict' object has no attribute 'func'
 2025-01-24 11:01:32 +08:00
Magic_yuan
443aab2882 修复当出现异常时，会导致更新数据卡死的bug 2025-01-24 10:15:25 +08:00
黄渝斌
3ea2ecbcbf Merge branch 'main' into redis_impl 2025-01-23 23:02:53 +08:00
hyb
ff71952c8c feat: 新增ini文件读取数据库配置方式，方便生产环境，修改Lightrag ainsert方法_add_doc_keys获取方式，原来只过滤存在的，但这会让失败的文档无法再次存储，新增--chunk_size和--chunk_overlap_size方便生产环境，新增llm_binding：openai-ollama 方便用openai的同时使用ollama embedding 2025-01-23 22:58:57 +08:00
Soichi Sumi
a3306c4633 add "azure_openai" in llm_binding and embedding_binding 2025-01-23 21:30:57 +09:00
zrguo
63bd066307 Merge pull request #628 from danielaskdd/Fix-parms-missing-for-openai 
Fix parms missing for openai
 2025-01-23 11:52:17 +08:00
yangdx
11b618d434 Fix parameters missing in app creation in case openai is selected 2025-01-23 01:15:48 +08:00
hyb
e08905b398 feat: 增加redis KV存储，增加openai+neo4j+milvus+redis的demo测试，新增lightrag.py: RedisKVStorage，新增requirements.txt:aioredis依赖 2025-01-22 16:42:13 +08:00
zrguo
4e5ca51e38 Update __init__.py 2025-01-21 10:29:52 +08:00
yangdx
8a05ff459f Add environment variable for Ollama model tag 2025-01-21 03:13:13 +08:00
yangdx
81a0a7cff0 fixed linting 2025-01-21 01:03:37 +08:00
yangdx
2e62cf1c66 Add document indexing during FastAPI startup, Fix docx package name in requirements 
- Initialize DocumentManager earlier
- Add lifespan context manager
- Scan and index documents on startup
- Fix docx package name in requirements
 2025-01-21 00:50:11 +08:00
zrguo
326057deeb fixed linting 2025-01-20 17:28:40 +08:00
zrguo
d9a22850c5 Merge pull request #608 from ParisNeo/main 
Temporary fix to introduced problem in ollama and lollm
 2025-01-20 17:23:28 +08:00
Saifeddine ALOUI
501610c603 Update llm.py 2025-01-20 10:04:17 +01:00
Saifeddine ALOUI
a557878e4e Update llm.py 2025-01-20 09:04:32 +01:00
Saifeddine ALOUI
e4945c9653 Update llm.py 2025-01-20 08:58:08 +01:00
Saifeddine ALOUI
f18f484a87 Update llm.py 2025-01-20 08:54:18 +01:00
yangdx
9ab719b605 Updated API documentation for LLM bindings and configurations. 2025-01-20 15:35:43 +08:00
yangdx
a76abf23f2 Updated OpenAI configuration and documentation. 
- Added OpenAI alike example in .env.example
- Updated README for OpenAI alike server
- Modified API key handling in lightrag_server.py
 2025-01-20 14:50:06 +08:00
zrguo
f8ba76f7b8 Merge branch 'main' into main 2025-01-20 12:26:01 +08:00
Saifeddine ALOUI
70425b0357 fixed linting 2025-01-20 00:26:28 +01:00
Saifeddine ALOUI
9cae05e1ff Fixed a bug introduced by a modification by someone else in azure_openai_complete (please make sure you test before commiting code) 
Added api_key to lollms, ollama, openai for both llm and embedding bindings allowing to use api key protected services.
 2025-01-19 23:24:37 +01:00
Saifeddine ALOUI
34c051b0b9 Merge branch 'HKUDS:main' into main 2025-01-19 10:40:21 +01:00
yangdx
347843d545 Use LLM_MODEL env var in Azure OpenAI function 
- Remove model parameter from azure_openai_complete (all LLM complete functions must have the same parameter structure)
- Use LLM_MODEL env var in Azure OpenAI function
- Comment out Lollms example in .env.example (duplication with Ollama example)
 2025-01-19 14:04:03 +08:00
yangdx
b50fdeccaa Merge branch 'main' into ollama-emulation 2025-01-19 13:32:07 +08:00
yangdx
b008c586ea Refactor service and script paths for LightRAG. 
- Renamed service file to `lightrag-server.service.example`
- Updated `ExecStart` path in service file
- Corrected README.md service file reference
- Moved `start_lightrag.sh.example` to root
- No functionality is changed
 2025-01-19 13:28:29 +08:00
zrguo
476e31f1e5 Merge pull request #600 from danielaskdd/ollama-emulation 
Migrate Ollama emulation features from lightrag_ollama to lightrag_server
 2025-01-19 12:58:36 +08:00
yangdx
ef73ed4291 Install Lightrag as a Linux Service (sample files and installation guide) 2025-01-19 12:53:13 +08:00
yangdx
a7b37652cf Add document scan API notes in API README.md 2025-01-19 12:24:46 +08:00
yangdx
3a227701b2 pre-commit run --all-files 2025-01-19 10:44:46 +08:00
yangdx
fb9a645f5e Update README with Ollama API and Open WebUI details 
- Add section on query mode selection
- Separate Ollama API and Open WebUI details
- Clarify query prefix usage
 2025-01-19 10:38:01 +08:00
yangdx
a78be2ab17 pre-commit run --all-files 2025-01-19 08:07:26 +08:00
yangdx
ea88981146 Update README.md for LightRAG Server 2025-01-19 06:45:32 +08:00
yangdx
853a9d2064 Updated API version to 1.0.3 
- Bumped API version to 1.0.3
- Fixed version reference in server code
 2025-01-19 06:06:17 +08:00
yangdx
387be31f09 Refactor embedding function initialization and remove start-server.sh 
- Simplified RAG initialization logic by deduplicating embedding function
- Removed start-server.sh script which is not needed
- No functional changes to the application
 2025-01-19 05:19:02 +08:00
yangdx
8ea179a98b Migrate Ollama API to lightrag_server.py 2025-01-19 04:44:30 +08:00
Nick French
df69d386c5 Fixes #596 - Hardcoded model deployment name in azure_openai_complete 
Fixes #596

Update `azure_openai_complete` function to accept a model parameter with a default value of 'gpt-4o-mini'.

* Modify the function signature of `azure_openai_complete` to include a `model` parameter with a default value of 'gpt-4o-mini'.
* Pass the `model` parameter to the `azure_openai_complete_if_cache` function instead of the hardcoded model name 'conversation-4o-mini'.

---

For more details, open the [Copilot Workspace session](https://copilot-workspace.githubnext.com/HKUDS/LightRAG/issues/596?shareId=XXXX-XXXX-XXXX-XXXX).
 2025-01-17 12:10:26 -05:00
Saifeddine ALOUI
35f04b51e6 Update lightrag_server.py 2025-01-17 11:18:45 +01:00
zrguo
28a84b2aa2 Merge pull request #592 from danielaskdd/yangdx 
Add Ollama compatible API server
 2025-01-17 14:29:31 +08:00
yangdx
fde0aa32c7 pre-commit run --all-files 2025-01-17 14:28:24 +08:00
yangdx
a561879040 Translate comments to English 2025-01-17 14:27:27 +08:00
yangdx
fa9765ecd9 pre-commit run --all-files 2025-01-17 14:20:55 +08:00
yangdx
939e399dd4 Translate comment to English 2025-01-17 13:36:31 +08:00