Add ability to use LLMs and embedding models from Ollama running loca…

…lly. (#45) * add new ollama provider, ability to use local ollama llama3.1 llm * revert header override for local ollama. * added in ollama nomic-embed-text embeddings * remove toggle to use ollama and fallback to using llm names. * add in ollama embeddings - mxbai-embed-large * configure ollama url via settings, and throw expection if not set. * adjust dimensions for mxbai_embed_large * retain bearer header incase needed for hosted ollama * updated with linting changes. * remove incept5/llama3.1-claude and be explicit on llama3.1:8b * removed usage of openAIKey for ollama * update filenames to be more meaningful
glowingjade · Oct 27, 2024 · 32c1c98 · 32c1c98
1 parent 454b55a
commit 32c1c98
Show file tree

Hide file tree

Showing 20 changed files with 857 additions and 119 deletions.
diff --git a/drizzle/0002_create_vector_data_nomic_embed_text.sql b/drizzle/0002_create_vector_data_nomic_embed_text.sql
@@ -0,0 +1,10 @@
+CREATE TABLE IF NOT EXISTS "vector_data_nomic_embed_text" (
+	"id" serial PRIMARY KEY NOT NULL,
+	"path" text NOT NULL,
+	"mtime" bigint NOT NULL,
+	"content" text NOT NULL,
+	"embedding" vector(768),
+	"metadata" jsonb NOT NULL
+);
+--> statement-breakpoint
+CREATE INDEX IF NOT EXISTS "embeddingIndex_nomic_embed_text" ON "vector_data_nomic_embed_text" USING hnsw ("embedding" vector_cosine_ops);
diff --git a/drizzle/0003_create_vector_data_mxbai_embed_large.sql b/drizzle/0003_create_vector_data_mxbai_embed_large.sql
@@ -0,0 +1,10 @@
+CREATE TABLE IF NOT EXISTS "vector_data_mxbai_embed_large" (
+	"id" serial PRIMARY KEY NOT NULL,
+	"path" text NOT NULL,
+	"mtime" bigint NOT NULL,
+	"content" text NOT NULL,
+	"embedding" vector(1024),
+	"metadata" jsonb NOT NULL
+);
+--> statement-breakpoint
+CREATE INDEX IF NOT EXISTS "embeddingIndex_mxbai_embed_large" ON "vector_data_mxbai_embed_large" USING hnsw ("embedding" vector_cosine_ops);
diff --git a/drizzle/meta/0002_snapshot.json b/drizzle/meta/0002_snapshot.json
@@ -0,0 +1,192 @@
+{
+  "id": "66d66503-35b5-4c19-8433-529426a63956",
+  "prevId": "1084161a-8de5-4452-ba8b-6ffc3549d411",
+  "version": "7",
+  "dialect": "postgresql",
+  "tables": {
+    "public.vector_data_text_embedding_3_small": {
+      "name": "vector_data_text_embedding_3_small",
+      "schema": "",
+      "columns": {
+        "id": {
+          "name": "id",
+          "type": "serial",
+          "primaryKey": true,
+          "notNull": true
+        },
+        "path": {
+          "name": "path",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "mtime": {
+          "name": "mtime",
+          "type": "bigint",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "content": {
+          "name": "content",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "embedding": {
+          "name": "embedding",
+          "type": "vector(1536)",
+          "primaryKey": false,
+          "notNull": false
+        },
+        "metadata": {
+          "name": "metadata",
+          "type": "jsonb",
+          "primaryKey": false,
+          "notNull": true
+        }
+      },
+      "indexes": {
+        "embeddingIndex_text_embedding_3_small": {
+          "name": "embeddingIndex_text_embedding_3_small",
+          "columns": [
+            {
+              "expression": "embedding",
+              "isExpression": false,
+              "asc": true,
+              "nulls": "last",
+              "opclass": "vector_cosine_ops"
+            }
+          ],
+          "isUnique": false,
+          "concurrently": false,
+          "method": "hnsw",
+          "with": {}
+        }
+      },
+      "foreignKeys": {},
+      "compositePrimaryKeys": {},
+      "uniqueConstraints": {},
+      "checkConstraints": {}
+    },
+    "public.vector_data_text_embedding_3_large": {
+      "name": "vector_data_text_embedding_3_large",
+      "schema": "",
+      "columns": {
+        "id": {
+          "name": "id",
+          "type": "serial",
+          "primaryKey": true,
+          "notNull": true
+        },
+        "path": {
+          "name": "path",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "mtime": {
+          "name": "mtime",
+          "type": "bigint",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "content": {
+          "name": "content",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "embedding": {
+          "name": "embedding",
+          "type": "vector(3072)",
+          "primaryKey": false,
+          "notNull": false
+        },
+        "metadata": {
+          "name": "metadata",
+          "type": "jsonb",
+          "primaryKey": false,
+          "notNull": true
+        }
+      },
+      "indexes": {},
+      "foreignKeys": {},
+      "compositePrimaryKeys": {},
+      "uniqueConstraints": {},
+      "checkConstraints": {}
+    },
+    "public.vector_data_nomic_embed_text": {
+      "name": "vector_data_nomic_embed_text",
+      "schema": "",
+      "columns": {
+        "id": {
+          "name": "id",
+          "type": "serial",
+          "primaryKey": true,
+          "notNull": true
+        },
+        "path": {
+          "name": "path",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "mtime": {
+          "name": "mtime",
+          "type": "bigint",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "content": {
+          "name": "content",
+          "type": "text",
+          "primaryKey": false,
+          "notNull": true
+        },
+        "embedding": {
+          "name": "embedding",
+          "type": "vector(768)",
+          "primaryKey": false,
+          "notNull": false
+        },
+        "metadata": {
+          "name": "metadata",
+          "type": "jsonb",
+          "primaryKey": false,
+          "notNull": true
+        }
+      },
+      "indexes": {
+        "embeddingIndex_nomic_embed_text": {
+          "name": "embeddingIndex_nomic_embed_text",
+          "columns": [
+            {
+              "expression": "embedding",
+              "isExpression": false,
+              "asc": true,
+              "nulls": "last",
+              "opclass": "vector_cosine_ops"
+            }
+          ],
+          "isUnique": false,
+          "concurrently": false,
+          "method": "hnsw",
+          "with": {}
+        }
+      },
+      "foreignKeys": {},
+      "compositePrimaryKeys": {},
+      "uniqueConstraints": {},
+      "checkConstraints": {}
+    }
+  },
+  "enums": {},
+  "schemas": {},
+  "sequences": {},
+  "views": {},
+  "_meta": {
+    "columns": {},
+    "schemas": {},
+    "tables": {}
+  }
+}