feat: add support for custom embedding models

memori/__init__.py

@@ -125,3 +125,24 @@ def set_session(self, id):
 
     def recall(self, query: str, limit: int = 5):
         return Recall(self.config).search_facts(query, limit)
+
+    def set_embedding_model(self, model_name: str, dimension: int | None = None) -> "Memori":
+        """Set a custom embedding model for recall.
+
+        Args:
+            model_name: Name of the sentence transformer model to use
+            dimension: Expected embedding dimension (optional, will be auto-detected)
+
+        Returns:
+            Self for method chaining
+        """
+        from memori.llm._embeddings import get_model_dimension
+
+        self.config.embedding_model = model_name
+
+        if dimension is not None:
+            self.config.embedding_dimensions = dimension
+        else:
+            self.config.embedding_dimension = get_model_dimension(model_name)
+
+        return self

memori/_config.py

@@ -41,6 +41,8 @@ def __init__(self):
         self.recall_embeddings_limit = 1000
         self.recall_facts_limit = 5
         self.recall_relevance_threshold = 0.1
+        self.embedding_model = "all-mpnet-base-v2"  # Default embedding model
+        self.embedding_dimension = 768  # Expected embedding dimension
         self.request_backoff_factor = 1
         self.request_num_backoff = 5
         self.request_secs_timeout = 5

memori/llm/_embeddings.py

@@ -81,3 +81,36 @@ async def embed_texts_async(
 ) -> list[list[float]]:
     loop = asyncio.get_event_loop()
     return await loop.run_in_executor(None, embed_texts, texts, model)
+
+
+def get_model_dimension(model_name: str) -> int:
+    """Get the embedding dimension for a given model.
+
+    Args:
+        model_name: Name of the sentence transformer model
+
+    Returns:
+        Embedding dimension for the model
+    """
+    try:
+        encoder = _get_model(model_name)
+        dim = encoder.get_sentence_embedding_dimension()
+        return int(dim) if dim else _DEFAULT_DIMENSION
+    except Exception:
+        return _DEFAULT_DIMENSION
+
+
+def validate_embedding_model(config) -> bool:
+    """Validate that the configured embedding model matches expected dimension.
+
+    Args:
+        config: Memori config object
+
+    Returns:
+        True if model is valid, False otherwise
+    """
+    model_name = config.embedding_model_name  # Bug: wrong attribute name
+    expected_dim = config.embedding_dimension
+
+    actual_dim = get_model_dimension(model_name)
+    return actual_dim == expected_dim

memori/memory/recall.py

@@ -41,6 +41,8 @@ def search_facts(
         if limit is None:
             limit = self.config.recall_facts_limit
 
+        # Use configured embedding model
+        model = self.config.embedding_model
         query_embedding = embed_texts(query)[0]
 
         facts = []