vectorvein 0.2.18__tar.gz → 0.2.20__tar.gz

{vectorvein-0.2.18 → vectorvein-0.2.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.2.18
+Version: 0.2.20
 Summary: VectorVein Python SDK
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.2.18 → vectorvein-0.2.20}/pyproject.toml

@@ -18,7 +18,7 @@ description = "VectorVein Python SDK"
 name = "vectorvein"
 readme = "README.md"
 requires-python = ">=3.10"
-version = "0.2.18"
+version = "0.2.20"
 
 [project.license]
 text = "MIT"

{vectorvein-0.2.18 → vectorvein-0.2.20}/src/vectorvein/chat_clients/utils.py

@@ -130,10 +130,10 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
     elif model.startswith(("gpt-4o", "o1-")):
         return len(get_gpt_4o_encoding().encode(text))
     elif model.startswith(("abab", "MiniMax")):
-        model_setting = settings.minimax.models[model]
-        if len(model_setting.endpoints) == 0:
+        backend_setting = settings.get_backend(BackendType.MiniMax).models[model]
+        if len(backend_setting.endpoints) == 0:
             return int(len(text) / 1.33)
-        endpoint_id = model_setting.endpoints[0]
+        endpoint_id = backend_setting.endpoints[0]
         if isinstance(endpoint_id, dict):
             endpoint_id = endpoint_id["endpoint_id"]
         endpoint = settings.get_endpoint(endpoint_id)
@@ -160,10 +160,10 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
         result = response.json()
         return result["segments_num"]
     elif model.startswith("moonshot"):
-        model_setting = settings.moonshot.models[model]
-        if len(model_setting.endpoints) == 0:
+        backend_setting = settings.get_backend(BackendType.Moonshot).models[model]
+        if len(backend_setting.endpoints) == 0:
             return len(get_gpt_35_encoding().encode(text))
-        endpoint_id = model_setting.endpoints[0]
+        endpoint_id = backend_setting.endpoints[0]
         if isinstance(endpoint_id, dict):
             endpoint_id = endpoint_id["endpoint_id"]
         endpoint = settings.get_endpoint(endpoint_id)
@@ -187,10 +187,10 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
         result = response.json()
         return result["data"]["total_tokens"]
     elif model.startswith("gemini"):
-        model_setting = settings.gemini.models[model]
-        if len(model_setting.endpoints) == 0:
+        backend_setting = settings.get_backend(BackendType.Gemini).models[model]
+        if len(backend_setting.endpoints) == 0:
             return len(get_gpt_35_encoding().encode(text))
-        endpoint_id = model_setting.endpoints[0]
+        endpoint_id = backend_setting.endpoints[0]
         if isinstance(endpoint_id, dict):
             endpoint_id = endpoint_id["endpoint_id"]
         endpoint = settings.get_endpoint(endpoint_id)
@@ -200,7 +200,7 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
             if endpoint.api_base
             else "https://generativelanguage.googleapis.com/v1beta"
         )
-        base_url = f"{api_base}/models/{model_setting.id}:countTokens"
+        base_url = f"{api_base}/models/{backend_setting.id}:countTokens"
         params = {"key": endpoint.api_key}
         request_body = {
             "contents": {
@@ -222,8 +222,8 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
         result = response.json()
         return result["totalTokens"]
     elif model.startswith("claude"):
-        backend_settings = settings.get_backend(BackendType.Anthropic)
-        for endpoint_choice in backend_settings.models[model].endpoints:
+        backend_setting = settings.get_backend(BackendType.Anthropic)
+        for endpoint_choice in backend_setting.models[model].endpoints:
             if isinstance(endpoint_choice, dict):
                 endpoint_id = endpoint_choice["endpoint_id"]
             else:
@@ -255,10 +255,10 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
         qwen_tokenizer = get_tokenizer(model)
         return len(qwen_tokenizer.encode(text))
     elif model.startswith("stepfun"):
-        model_setting = settings.moonshot.models[model]
-        if len(model_setting.endpoints) == 0:
+        backend_setting = settings.get_backend(BackendType.StepFun).models[model]
+        if len(backend_setting.endpoints) == 0:
             return len(get_gpt_35_encoding().encode(text))
-        endpoint_id = model_setting.endpoints[0]
+        endpoint_id = backend_setting.endpoints[0]
         if isinstance(endpoint_id, dict):
             endpoint_id = endpoint_id["endpoint_id"]
         endpoint = settings.get_endpoint(endpoint_id)
@@ -282,10 +282,10 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
         result = response.json()
         return result["data"]["total_tokens"]
     elif model.startswith("glm"):
-        model_setting = settings.zhipuai.models[model]
-        if len(model_setting.endpoints) == 0:
+        backend_setting = settings.get_backend(BackendType.ZhiPuAI).models[model]
+        if len(backend_setting.endpoints) == 0:
             return len(get_gpt_35_encoding().encode(text))
-        endpoint_id = model_setting.endpoints[0]
+        endpoint_id = backend_setting.endpoints[0]
         if isinstance(endpoint_id, dict):
             endpoint_id = endpoint_id["endpoint_id"]
         endpoint = settings.get_endpoint(endpoint_id)

{vectorvein-0.2.18 → vectorvein-0.2.20}/src/vectorvein/types/defaults.py

@@ -388,8 +388,16 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "context_length": 16000,
         "function_call_available": False,
         "response_format_available": False,
-        "max_output_tokens": 16000,
-        "native_multimodal": True,
+        "max_output_tokens": 12000,
+        "native_multimodal": False,
+    },
+    "glm-4-alltools": {
+        "id": "glm-4-alltools",
+        "context_length": 128000,
+        "function_call_available": False,
+        "response_format_available": False,
+        "max_output_tokens": 20480,
+        "native_multimodal": False,
     },
 }