gmakstutis/LightRAG
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update summary_max_tokens default from 10k to 30k tokens

Browse source
This commit is contained in:
yangdx 2025-08-21 23:16:07 +08:00
parent dce678642a
commit 16a1ef1178
4 changed files with 4 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
README-zh.md
View file

@ -267,7 +267,7 @@ if __name__ == "__main__":
| **embedding_func_max_async** | `int` | 最大并发异步嵌入进程数 | `16` |
| **llm_model_func** | `callable` | LLM生成的函数 | `gpt_4o_mini_complete` |
| **llm_model_name** | `str` | 用于生成的LLM模型名称 | `meta-llama/Llama-3.2-1B-Instruct` |
| **summary_max_tokens** | `int` | 生成实体关系摘要时送给LLM的最大令牌数 | `32000`（由环境变量 SUMMARY_MAX_TOKENS 设置） |
| **summary_max_tokens** | `int` | 生成实体关系摘要时送给LLM的最大令牌数 | `30000`（由环境变量 SUMMARY_MAX_TOKENS 设置） |
| **llm_model_max_async** | `int` | 最大并发异步LLM进程数 | `4`（默认值由环境变量MAX_ASYNC更改） |
| **llm_model_kwargs** | `dict` | LLM生成的附加参数 | |
| **vector_db_storage_cls_kwargs** | `dict` | 向量数据库的附加参数，如设置节点和关系检索的阈值 | cosine_better_than_threshold: 0.2（默认值由环境变量COSINE_THRESHOLD更改） |

2
README.md
View file

@ -274,7 +274,7 @@ A full list of LightRAG init parameters:
| **embedding_func_max_async** | `int` | Maximum number of concurrent asynchronous embedding processes | `16` |
| **llm_model_func** | `callable` | Function for LLM generation | `gpt_4o_mini_complete` |
| **llm_model_name** | `str` | LLM model name for generation | `meta-llama/Llama-3.2-1B-Instruct` |
| **summary_max_tokens** | `int` | Maximum tokens send to LLM to generate entity relation summaries | `32000`（configured by env var SUMMARY_MAX_TOKENS) |
| **summary_max_tokens** | `int` | Maximum tokens send to LLM to generate entity relation summaries | `30000`（configured by env var SUMMARY_MAX_TOKENS) |
| **llm_model_max_async** | `int` | Maximum number of concurrent asynchronous LLM processes | `4`（default value changed by env var MAX_ASYNC) |
| **llm_model_kwargs** | `dict` | Additional parameters for LLM generation | |
| **vector_db_storage_cls_kwargs** | `dict` | Additional parameters for vector database, like setting the threshold for nodes and relations retrieval | cosine_better_than_threshold: 0.2（default value changed by env var COSINE_THRESHOLD) |

2
env.example
View file

@ -107,7 +107,7 @@ ENABLE_LLM_CACHE_FOR_EXTRACT=true
### Entity and relation summarization configuration
### Number of duplicated entities/edges to trigger LLM re-summary on merge (at least 3 is recommented)， and max tokens send to LLM
# FORCE_LLM_SUMMARY_ON_MERGE=4
# SUMMARY_MAX_TOKENS=10000
# SUMMARY_MAX_TOKENS=30000
### Maximum number of entity extraction attempts for ambiguous content
# MAX_GLEANING=1

2
lightrag/constants.py
View file

@ -14,7 +14,7 @@ DEFAULT_MAX_GRAPH_NODES = 1000
DEFAULT_SUMMARY_LANGUAGE = "English" # Default language for summaries
DEFAULT_FORCE_LLM_SUMMARY_ON_MERGE = 4
DEFAULT_MAX_GLEANING = 1
DEFAULT_SUMMARY_MAX_TOKENS = 10000 # Default maximum token size
DEFAULT_SUMMARY_MAX_TOKENS = 30000 # Default maximum token size
# Separator for graph fields
GRAPH_FIELD_SEP = "<SEP>"