ryan/simbarag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Switch to OpenAI embeddings for ChromaDB

Browse Source 
Replace Ollama embedding function with OpenAI's text-embedding-3-small
model for improved embedding quality and consistency.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Ryan Chen
2025-10-02 21:05:17 -04:00
parent c5091dc07a
commit 3ffc95a1b0
1 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
chunker.py
View File

@@ -4,8 +4,8 @@ import re
from typing import Union from typing import Union
from uuid import UUID, uuid4 from uuid import UUID, uuid4
from chromadb.utils.embedding_functions.ollama_embedding_function import ( from chromadb.utils.embedding_functions.openai_embedding_function import (
OllamaEmbeddingFunction, OpenAIEmbeddingFunction,
) )
from dotenv import load_dotenv from dotenv import load_dotenv
@@ -80,9 +80,9 @@ class Chunk:
class Chunker: class Chunker:
embedding_fx = OllamaEmbeddingFunction( embedding_fx = OpenAIEmbeddingFunction(
url=os.getenv("OLLAMA_URL", "http://host.docker.internal:11434"), api_key=os.getenv("OPENAI_API_KEY"),
model_name="mxbai-embed-large", model_name="text-embedding-3-small",
) )
def __init__(self, collection) -> None: def __init__(self, collection) -> None: