From 40e5e96212da183883ef6904c87bdfd84a445f15 Mon Sep 17 00:00:00 2001
From: n0x29a <n0x29a@example.com>
Date: Tue, 18 Feb 2025 17:57:33 +0100
Subject: [PATCH] draft

---
 fastembed/config/dense_models.json   | 16 ++++++++++++++++
 fastembed/config/sparse_models.json  |  0
 fastembed/model_loader.py            | 16 ++++++++++++++++
 3 files changed, 32 insertions(+)
 create mode 100644 fastembed/config/dense_models.json
 create mode 100644 fastembed/config/sparse_models.json 
 create mode 100644 fastembed/model_loader.py

diff --git a/fastembed/config/dense_models.json b/fastembed/config/dense_models.json
new file mode 100644
index 0000000..1d6fad8
--- /dev/null
+++ b/fastembed/config/dense_models.json
@@ -0,0 +1,16 @@
+{
+  "models": [
+    {
+      "model": "BAAI/bge-base-en",
+      "dim": 768,
+      "description": "Text embeddings, Unimodal (text), English...",
+      "license": "mit",
+      "size_in_GB": 0.42,
+      "sources": {
+        "hf": "Qdrant/fast-bge-base-en",
+        "url": "https://storage.googleapis.com/qdrant-fastembed/fast-bge-base-en.tar.gz"
+      },
+      "model_file": "model_optimized.onnx"
+    }
+  ]
+}
diff --git a/fastembed/config/sparse_models.json  b/fastembed/config/sparse_models.json 
new file mode 100644
index 0000000..e69de29
diff --git a/fastembed/model_loader.py b/fastembed/model_loader.py
new file mode 100644
index 0000000..e6ff9b3
--- /dev/null
+++ b/fastembed/model_loader.py
@@ -0,0 +1,16 @@
+from pathlib import Path
+import json
+from typing import Dict, List
+
+
+class ModelLoader:
+    def __init__(self):
+        self.config_dir = Path(__file__).parent / "configs"
+        self._models: Dict[str, List[Dict]] = {}
+
+    def load_models(self, model_type: str) -> List[Dict]:
+        if model_type not in self._models:
+            config_path = self.config_dir / f"{model_type}_models.json"
+            with open(config_path) as f:
+                self._models[model_type] = json.load(f)["models"]
+        return self._models[model_type]