Disable automatic extraction and deduplication for local models (#1831)

* Disable automatic extraction and deduplication for local

* Bump js

js/sdk/package.json

@@ -1,6 +1,6 @@
 {
   "name": "r2r-js",
-  "version": "0.4.16",
+  "version": "0.4.17",
   "description": "",
   "main": "dist/index.js",
   "browser": "dist/index.browser.js",

py/core/configs/full_lm_studio.toml

@@ -31,6 +31,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "lm_studio/llama-3.2-3b-instruct" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -60,6 +61,7 @@ combine_under_n_chars = 128
 overlap = 20
 chunks_for_document_summary = 16
 document_summary_model = "lm_studio/llama-3.2-3b-instruct"
+automatic_extraction = false
 
 [orchestration]
 provider = "hatchet"

py/core/configs/full_local_llm.toml

@@ -31,6 +31,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "ollama/llama3.1" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -60,6 +61,7 @@ combine_under_n_chars = 128
 overlap = 20
 chunks_for_document_summary = 16
 document_summary_model = "ollama/llama3.1"
+automatic_extraction = false
 
 [orchestration]
 provider = "hatchet"

py/core/configs/full_ollama.toml

@@ -31,6 +31,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "ollama/llama3.1" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -60,6 +61,7 @@ combine_under_n_chars = 128
 overlap = 20
 chunks_for_document_summary = 16
 document_summary_model = "ollama/llama3.1"
+automatic_extraction = false
 
 [orchestration]
 provider = "hatchet"

py/core/configs/lm_studio.toml

@@ -37,6 +37,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "lm_studio/llama-3.2-3b-instruct" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -57,6 +58,7 @@ vision_img_model = "lm_studio/llama3.2-vision"
 vision_pdf_model = "lm_studio/llama3.2-vision"
 chunks_for_document_summary = 16
 document_summary_model = "lm_studio/llama-3.2-3b-instruct"
+automatic_extraction = false
 
   [ingestion.extra_parsers]
     pdf = "zerox"

py/core/configs/local_llm.toml

@@ -37,6 +37,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "ollama/llama3.1" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -57,6 +58,7 @@ vision_img_model = "ollama/llama3.2-vision"
 vision_pdf_model = "ollama/llama3.2-vision"
 chunks_for_document_summary = 16
 document_summary_model = "ollama/llama3.1"
+automatic_extraction = false
 
   [ingestion.extra_parsers]
     pdf = "zerox"

py/core/configs/ollama.toml

@@ -37,6 +37,7 @@ provider = "postgres"
     max_knowledge_relationships = 100
     max_description_input_length = 65536
     generation_config = { model = "ollama/llama3.1" } # and other params, model used for relationshipt extraction
+    automatic_deduplication = false
 
   [database.graph_enrichment_settings]
     community_reports_prompt = "graphrag_community_reports"
@@ -57,6 +58,7 @@ vision_img_model = "ollama/llama3.2-vision"
 vision_pdf_model = "ollama/llama3.2-vision"
 chunks_for_document_summary = 16
 document_summary_model = "ollama/llama3.1"
+automatic_extraction = false
 
   [ingestion.extra_parsers]
     pdf = "zerox"