From 2ffd67bf0de11b483adff9965ff28e527d55581b Mon Sep 17 00:00:00 2001
From: Tommy Smith <tommy@weaviate.io>
Date: Wed, 18 Sep 2024 10:44:25 +0100
Subject: [PATCH 1/2] Add `baseURL` and `dimensions` params to
 `text2vec-jinaai`

---
 .../classes/config_named_vectors.py           |  8 ++++++
 .../collections/classes/config_vectorizers.py | 26 +++++++++++++++++--
 2 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/weaviate/collections/classes/config_named_vectors.py b/weaviate/collections/classes/config_named_vectors.py
index fcb08fcac..c5b45fcd1 100644
--- a/weaviate/collections/classes/config_named_vectors.py
+++ b/weaviate/collections/classes/config_named_vectors.py
@@ -951,6 +951,8 @@ def text2vec_jinaai(
         source_properties: Optional[List[str]] = None,
         vector_index_config: Optional[_VectorIndexConfigCreate] = None,
         vectorize_collection_name: bool = True,
+        base_url: Optional[str] = None,
+        dimensions: Optional[int] = None,
         model: Optional[Union[JinaModel, str]] = None,
     ) -> _NamedVectorConfigCreate:
         """Create a named vector using the `text2vec-jinaai` model.
@@ -967,6 +969,10 @@ def text2vec_jinaai(
                 The configuration for Weaviate's vector index. Use wvc.config.Configure.VectorIndex to create a vector index configuration. None by default
             `vectorize_collection_name`
                 Whether to vectorize the collection name. Defaults to `True`.
+            `base_url`
+                The base URL to send the vectorization requests to. Defaults to `None`, which uses the server-defined default.
+            `dimensions`
+                The number of dimensions for the generated embeddings. Defaults to `None`, which uses the server-defined default.
             `model`
                 The model to use. Defaults to `None`, which uses the server-defined default.
                 See the
@@ -976,6 +982,8 @@ def text2vec_jinaai(
             name=name,
             source_properties=source_properties,
             vectorizer=_Text2VecJinaConfigCreate(
+                baseURL=base_url,
+                dimensions=dimensions,
                 model=model,
                 vectorizeClassName=vectorize_collection_name,
             ),
diff --git a/weaviate/collections/classes/config_vectorizers.py b/weaviate/collections/classes/config_vectorizers.py
index 3e8a1a018..b4e7c466d 100644
--- a/weaviate/collections/classes/config_vectorizers.py
+++ b/weaviate/collections/classes/config_vectorizers.py
@@ -24,7 +24,16 @@
 OpenAIModel: TypeAlias = Literal[
     "text-embedding-3-small", "text-embedding-3-large", "text-embedding-ada-002"
 ]
-JinaModel: TypeAlias = Literal["jina-embeddings-v2-base-en", "jina-embeddings-v2-small-en"]
+JinaModel: TypeAlias = Literal[
+    "jina-embeddings-v2-base-en",
+    "jina-embeddings-v2-small-en",
+    "jina-embeddings-v2-base-zh",
+    "jina-embeddings-v2-base-es",
+    "jina-embeddings-v2-base-code",
+    "jina-colbert-v1-en",
+    "jina-colbert-v2",
+    "jina-embeddings-v3",
+]
 VoyageModel: TypeAlias = Literal[
     "voyage-large-2",
     "voyage-code-2",
@@ -341,6 +350,8 @@ class _Text2VecJinaConfig(_ConfigCreateModel):
     vectorizer: Union[Vectorizers, _EnumLikeStr] = Field(
         default=Vectorizers.TEXT2VEC_JINAAI, frozen=True, exclude=True
     )
+    baseURL: Optional[str]
+    dimensions: Optional[int]
     model: Optional[str]
     vectorizeClassName: bool
 
@@ -1094,6 +1105,8 @@ def text2vec_transformers(
     def text2vec_jinaai(
         model: Optional[Union[JinaModel, str]] = None,
         vectorize_collection_name: bool = True,
+        base_url: Optional[str] = None,
+        dimensions: Optional[int] = None,
     ) -> _VectorizerConfigCreate:
         """Create a `_Text2VecJinaConfigCreate` object for use when vectorizing using the `text2vec-jinaai` model.
 
@@ -1107,8 +1120,17 @@ def text2vec_jinaai(
                 [documentation](https://weaviate.io/developers/weaviate/modules/retriever-vectorizer-modules/text2vec-jinaai#available-models) for more details.
             `vectorize_collection_name`
                 Whether to vectorize the collection name. Defaults to `True`.
+            `base_url`
+                The base URL to send the vectorization requests to. Defaults to `None`, which uses the server-defined default.
+            `dimensions`
+                The number of dimensions for the generated embeddings. Defaults to `None`, which uses the server-defined default.
         """
-        return _Text2VecJinaConfigCreate(model=model, vectorizeClassName=vectorize_collection_name)
+        return _Text2VecJinaConfigCreate(
+            model=model,
+            vectorizeClassName=vectorize_collection_name,
+            baseURL=base_url,
+            dimensions=dimensions,
+        )
 
     @staticmethod
     def text2vec_voyageai(

From 3a7e1826d6082b4041c57e49a0f0cdba2a3a660e Mon Sep 17 00:00:00 2001
From: Tommy Smith <tommy@weaviate.io>
Date: Wed, 18 Sep 2024 10:49:06 +0100
Subject: [PATCH 2/2] Remove colbert from allowed JinaModels

---
 weaviate/collections/classes/config_vectorizers.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/weaviate/collections/classes/config_vectorizers.py b/weaviate/collections/classes/config_vectorizers.py
index b4e7c466d..056c73d73 100644
--- a/weaviate/collections/classes/config_vectorizers.py
+++ b/weaviate/collections/classes/config_vectorizers.py
@@ -30,8 +30,6 @@
     "jina-embeddings-v2-base-zh",
     "jina-embeddings-v2-base-es",
     "jina-embeddings-v2-base-code",
-    "jina-colbert-v1-en",
-    "jina-colbert-v2",
     "jina-embeddings-v3",
 ]
 VoyageModel: TypeAlias = Literal[