vectorvein 0.1.18__tar.gz → 0.1.19__tar.gz

{vectorvein-0.1.18 → vectorvein-0.1.19}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.1.18
+Version: 0.1.19
 Summary: Default template for PDM package
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.1.18 → vectorvein-0.1.19}/pyproject.toml

@@ -16,7 +16,7 @@ description = "Default template for PDM package"
 name = "vectorvein"
 readme = "README.md"
 requires-python = ">=3.10"
-version = "0.1.18"
+version = "0.1.19"
 
 [project.license]
 text = "MIT"

{vectorvein-0.1.18 → vectorvein-0.1.19}/src/vectorvein/chat_clients/openai_compatible_client.py

@@ -90,25 +90,6 @@ class OpenAICompatibleChatClient(BaseChatClient):
 
         self.model_setting = self.backend_settings.models[self.model]
 
-        # if self.random_endpoint:
-        #     self.random_endpoint = True
-        #     self.endpoint_id = random.choice(self.backend_settings.models[self.model].endpoints)
-        #     self.endpoint = settings.get_endpoint(self.endpoint_id)
-
-        # if self.endpoint.is_azure:
-        #     self._client = AzureOpenAI(
-        #         azure_endpoint=self.endpoint.api_base,
-        #         api_key=self.endpoint.api_key,
-        #         api_version="2024-08-01-preview",
-        #         http_client=self.http_client,
-        #     )
-        # else:
-        #     self._client = OpenAI(
-        #         api_key=self.endpoint.api_key,
-        #         base_url=self.endpoint.api_base,
-        #         http_client=self.http_client,
-        #     )
-
         if self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,
@@ -269,25 +250,6 @@ class AsyncOpenAICompatibleChatClient(BaseAsyncChatClient):
 
         self.model_setting = self.backend_settings.models[self.model]
 
-        # if self.random_endpoint:
-        #     self.random_endpoint = True
-        #     self.endpoint_id = random.choice(self.backend_settings.models[self.model].endpoints)
-        #     self.endpoint = settings.get_endpoint(self.endpoint_id)
-
-        # if self.endpoint.is_azure:
-        #     self._client = AsyncAzureOpenAI(
-        #         azure_endpoint=self.endpoint.api_base,
-        #         api_key=self.endpoint.api_key,
-        #         api_version="2024-08-01-preview",
-        #         http_client=self.http_client,
-        #     )
-        # else:
-        #     self._client = AsyncOpenAI(
-        #         api_key=self.endpoint.api_key,
-        #         base_url=self.endpoint.api_base,
-        #         http_client=self.http_client,
-        #     )
-
         if self.context_length_control == ContextLengthControlType.Latest:
             messages = cutoff_messages(
                 messages,

{vectorvein-0.1.18 → vectorvein-0.1.19}/src/vectorvein/types/defaults.py

@@ -119,6 +119,34 @@ GROQ_MODELS = {
         "function_call_available": True,
         "response_format_available": True,
     },
+    "gemma2-9b-it": {
+        "id": "gemma2-9b-it",
+        "context_length": 8192,
+    },
+    "llama3-groq-70b-8192-tool-use-preview": {
+        "id": "llama3-groq-70b-8192-tool-use-preview",
+        "context_length": 8192,
+        "function_call_available": True,
+        "max_output_tokens": 8000,
+    },
+    "llama3-groq-8b-8192-tool-use-preview": {
+        "id": "llama3-groq-8b-8192-tool-use-preview",
+        "context_length": 8192,
+        "function_call_available": True,
+        "max_output_tokens": 8000,
+    },
+    "llama-3.1-70b-versatile": {
+        "id": "llama-3.1-70b-versatile",
+        "context_length": 131072,
+        "function_call_available": True,
+        "max_output_tokens": 8000,
+    },
+    "llama-3.1-8b-instant": {
+        "id": "llama-3.1-8b-instant",
+        "context_length": 131072,
+        "function_call_available": True,
+        "max_output_tokens": 8000,
+    },
 }
 
 # Qwen models