vectorvein 0.2.76__tar.gz → 0.2.77__tar.gz

{vectorvein-0.2.76 → vectorvein-0.2.77}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectorvein
-Version: 0.2.76
+Version: 0.2.77
 Summary: VectorVein Python SDK
 Author-Email: Anderson <andersonby@163.com>
 License: MIT

{vectorvein-0.2.76 → vectorvein-0.2.77}/pyproject.toml

@@ -17,7 +17,7 @@ description = "VectorVein Python SDK"
 name = "vectorvein"
 readme = "README.md"
 requires-python = ">=3.10"
-version = "0.2.76"
+version = "0.2.77"
 
 [project.license]
 text = "MIT"

{vectorvein-0.2.76 → vectorvein-0.2.77}/src/vectorvein/chat_clients/utils.py

@@ -158,7 +158,7 @@ def get_token_counts(text: str | dict, model: str = "", use_token_server_first:
             return 1000
         result = response.json()
         return result["segments_num"]
-    elif model.startswith("moonshot"):
+    elif model.startswith(("moonshot", "kimi")):
         backend_setting = settings.get_backend(BackendType.Moonshot).models[model]
         if len(backend_setting.endpoints) == 0:
             return len(get_gpt_35_encoding().encode(text))

{vectorvein-0.2.76 → vectorvein-0.2.77}/src/vectorvein/types/defaults.py

@@ -1,6 +1,6 @@
 # @Author: Bi Ying
 # @Date:   2024-07-27 00:02:34
-from typing import Final, Dict, Any
+from typing import Final, Dict, TypedDict, NotRequired
 
 from .enums import ContextLengthControlType
 
@@ -12,26 +12,39 @@ ENDPOINT_TPM: Final[int] = 300000
 
 MODEL_CONTEXT_LENGTH: Final[int] = 32768
 
+
+class ModelSettingDict(TypedDict):
+    id: str
+    function_call_available: bool
+    response_format_available: bool
+    native_multimodal: bool
+    context_length: int
+    max_output_tokens: NotRequired[int]
+
+
 # Moonshot models
-MOONSHOT_DEFAULT_MODEL: Final[str] = "moonshot-v1-8k"
-MOONSHOT_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+MOONSHOT_DEFAULT_MODEL: Final[str] = "kimi-latest"
+MOONSHOT_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "moonshot-v1-8k": {
         "id": "moonshot-v1-8k",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "moonshot-v1-32k": {
         "id": "moonshot-v1-32k",
         "context_length": 32768,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "moonshot-v1-128k": {
         "id": "moonshot-v1-128k",
         "context_length": 131072,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "moonshot-v1-8k-vision-preview": {
         "id": "moonshot-v1-8k-vision-preview",
@@ -55,17 +68,25 @@ MOONSHOT_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "response_format_available": True,
         "native_multimodal": True,
     },
+    "kimi-latest": {
+        "id": "kimi-latest",
+        "context_length": 131072,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": True,
+    },
 }
 
 # Deepseek models
 DEEPSEEK_DEFAULT_MODEL: Final[str] = "deepseek-chat"
-DEEPSEEK_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+DEEPSEEK_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "deepseek-chat": {
         "id": "deepseek-chat",
         "context_length": 64000,
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "deepseek-reasoner": {
         "id": "deepseek-reasoner",
@@ -73,18 +94,20 @@ DEEPSEEK_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
 }
 
 # Baichuan models
 BAICHUAN_DEFAULT_MODEL: Final[str] = "Baichuan3-Turbo"
-BAICHUAN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+BAICHUAN_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "Baichuan4": {
         "id": "Baichuan4",
         "context_length": 32768,
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "Baichuan3-Turbo": {
         "id": "Baichuan3-Turbo",
@@ -92,6 +115,7 @@ BAICHUAN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "Baichuan3-Turbo-128k": {
         "id": "Baichuan3-Turbo-128k",
@@ -99,6 +123,7 @@ BAICHUAN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "Baichuan2-Turbo": {
         "id": "Baichuan2-Turbo",
@@ -106,6 +131,7 @@ BAICHUAN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "Baichuan2-53B": {
         "id": "Baichuan2-53B",
@@ -113,75 +139,92 @@ BAICHUAN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
 }
 
 # Groq models
 GROQ_DEFAULT_MODEL: Final[str] = "llama3-70b-8192"
-GROQ_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+GROQ_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "mixtral-8x7b-32768": {
         "id": "mixtral-8x7b-32768",
         "context_length": 32768,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "llama3-70b-8192": {
         "id": "llama3-70b-8192",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "llama3-8b-8192": {
         "id": "llama3-8b-8192",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "gemma-7b-it": {
         "id": "gemma-7b-it",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "gemma2-9b-it": {
         "id": "gemma2-9b-it",
         "context_length": 8192,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": False,
     },
     "llama3-groq-70b-8192-tool-use-preview": {
         "id": "llama3-groq-70b-8192-tool-use-preview",
         "context_length": 8192,
         "function_call_available": True,
+        "response_format_available": True,
         "max_output_tokens": 8000,
+        "native_multimodal": False,
     },
     "llama3-groq-8b-8192-tool-use-preview": {
         "id": "llama3-groq-8b-8192-tool-use-preview",
         "context_length": 8192,
         "function_call_available": True,
+        "response_format_available": True,
         "max_output_tokens": 8000,
+        "native_multimodal": False,
     },
     "llama-3.1-70b-versatile": {
         "id": "llama-3.1-70b-versatile",
         "context_length": 131072,
         "function_call_available": True,
+        "response_format_available": True,
         "max_output_tokens": 8000,
+        "native_multimodal": False,
     },
     "llama-3.1-8b-instant": {
         "id": "llama-3.1-8b-instant",
         "context_length": 131072,
         "function_call_available": True,
+        "response_format_available": True,
         "max_output_tokens": 8000,
+        "native_multimodal": False,
     },
 }
 
 # Qwen models
 QWEN_DEFAULT_MODEL: Final[str] = "qwen2.5-72b-instruct"
-QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+QWEN_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "qwen2.5-7b-instruct": {
         "id": "qwen2.5-7b-instruct",
         "context_length": 131072,
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen2.5-14b-instruct": {
         "id": "qwen2.5-14b-instruct",
@@ -189,6 +232,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen2.5-32b-instruct": {
         "id": "qwen2.5-32b-instruct",
@@ -196,6 +240,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen2.5-coder-32b-instruct": {
         "id": "qwen2.5-coder-32b-instruct",
@@ -203,6 +248,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "qwq-32b": {
         "id": "qwq-32b",
@@ -210,6 +256,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "qwen2.5-72b-instruct": {
         "id": "qwen2.5-72b-instruct",
@@ -217,6 +264,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen2-vl-72b-instruct": {
         "id": "qwen2-vl-72b-instruct",
@@ -256,6 +304,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen-max-longcontext": {
         "id": "qwen-max-longcontext",
@@ -263,6 +312,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen-plus": {
         "id": "qwen-plus",
@@ -270,6 +320,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8096,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qwen-turbo": {
         "id": "qwen-turbo",
@@ -277,6 +328,7 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 1500,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "qvq-max": {
         "id": "qvq-max",
@@ -290,13 +342,14 @@ QWEN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 # Yi models
 YI_DEFAULT_MODEL: Final[str] = "yi-lightning"
-YI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+YI_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "yi-lightning": {
         "id": "yi-lightning",
         "context_length": 16000,
         "max_output_tokens": 4096,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "yi-vision-v2": {
         "id": "yi-vision-v2",
@@ -310,13 +363,14 @@ YI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 # ZhiPuAI models
 ZHIPUAI_DEFAULT_MODEL: Final[str] = "glm-4-air"
-ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+ZHIPUAI_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "glm-3-turbo": {
         "id": "glm-3-turbo",
         "context_length": 128000,
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4": {
         "id": "glm-4",
@@ -324,6 +378,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-plus": {
         "id": "glm-4-plus",
@@ -331,6 +386,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-0520": {
         "id": "glm-4-0520",
@@ -338,6 +394,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-air": {
         "id": "glm-4-air",
@@ -345,6 +402,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-airx": {
         "id": "glm-4-airx",
@@ -352,6 +410,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-flash": {
         "id": "glm-4-flash",
@@ -359,6 +418,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4-long": {
         "id": "glm-4-long",
@@ -366,6 +426,7 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "function_call_available": True,
         "response_format_available": False,
         "max_output_tokens": 4095,
+        "native_multimodal": False,
     },
     "glm-4v": {
         "id": "glm-4v",
@@ -411,66 +472,75 @@ ZHIPUAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 # Mistral models
 MISTRAL_DEFAULT_MODEL: Final[str] = "mistral-small"
-MISTRAL_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+MISTRAL_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "open-mistral-7b": {
         "id": "open-mistral-7b",
         "context_length": 32000,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "open-mixtral-8x7b": {
         "id": "open-mixtral-8x7b",
         "context_length": 32000,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "open-mixtral-8x22b": {
         "id": "open-mixtral-8x22b",
         "context_length": 64000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "open-mistral-nemo": {
         "id": "open-mistral-nemo",
         "context_length": 128000,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "codestral-latest": {
         "id": "codestral-latest",
         "context_length": 32000,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "mistral-small-latest": {
         "id": "mistral-small-latest",
         "context_length": 30000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "mistral-medium-latest": {
         "id": "mistral-medium-latest",
         "context_length": 30000,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "mistral-large-latest": {
         "id": "mistral-large-latest",
         "context_length": 128000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
 }
 
 # OpenAI models
 OPENAI_DEFAULT_MODEL: Final[str] = "gpt-4o"
-OPENAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+OPENAI_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "gpt-35-turbo": {
         "id": "gpt-35-turbo",
         "context_length": 16385,
         "function_call_available": True,
         "response_format_available": True,
         "max_output_tokens": 4096,
+        "native_multimodal": False,
     },
     "gpt-4-turbo": {
         "id": "gpt-4-turbo",
@@ -478,6 +548,7 @@ OPENAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "gpt-4": {
         "id": "gpt-4",
@@ -485,6 +556,7 @@ OPENAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "gpt-4o": {
         "id": "gpt-4o",
@@ -508,6 +580,7 @@ OPENAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": True,
     },
     "o1-mini": {
         "id": "o1-mini",
@@ -536,8 +609,8 @@ OPENAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 }
 
 # Anthropic models
-ANTHROPIC_DEFAULT_MODEL: Final[str] = "claude-3-5-sonnet-20241022"
-ANTHROPIC_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+ANTHROPIC_DEFAULT_MODEL: Final[str] = "claude-3-7-sonnet-20250219"
+ANTHROPIC_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "claude-3-opus-20240229": {
         "id": "claude-3-opus-20240229",
         "context_length": 200000,
@@ -598,13 +671,14 @@ ANTHROPIC_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 # Minimax models
 MINIMAX_DEFAULT_MODEL: Final[str] = "MiniMax-Text-01"
-MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+MINIMAX_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "abab5-chat": {
         "id": "abab5-chat",
         "context_length": 6144,
         "max_output_tokens": 6144,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "abab5.5-chat": {
         "id": "abab5.5-chat",
@@ -612,6 +686,7 @@ MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 16384,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "abab6-chat": {
         "id": "abab6-chat",
@@ -619,6 +694,7 @@ MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 32768,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "abab6.5s-chat": {
         "id": "abab6.5s-chat",
@@ -626,6 +702,7 @@ MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 245760,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "abab7-preview": {
         "id": "abab7-preview",
@@ -633,6 +710,7 @@ MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 245760,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "MiniMax-Text-01": {
         "id": "MiniMax-Text-01",
@@ -640,12 +718,13 @@ MINIMAX_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 1000192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
 }
 
 # Gemini models
 GEMINI_DEFAULT_MODEL: Final[str] = "gemini-2.0-flash"
-GEMINI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+GEMINI_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "gemini-1.5-pro": {
         "id": "gemini-1.5-pro",
         "context_length": 2097152,
@@ -737,13 +816,14 @@ GEMINI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 # 百度文心一言 ERNIE 模型
 ERNIE_DEFAULT_MODEL: Final[str] = "ernie-lite"
-ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+ERNIE_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "ernie-lite": {
         "id": "ernie-lite",
         "context_length": 6144,
         "max_output_tokens": 2048,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "ernie-speed": {
         "id": "ernie-speed",
@@ -751,6 +831,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "ernie-speed-pro-128k": {
         "id": "ernie-speed-pro-128k",
@@ -758,6 +839,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 4096,
         "function_call_available": False,
         "response_format_available": False,
+        "native_multimodal": False,
     },
     "ernie-3.5-8k": {
         "id": "ernie-3.5-8k",
@@ -765,6 +847,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "ernie-3.5-128k": {
         "id": "ernie-3.5-128k",
@@ -772,6 +855,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "ernie-4.0-8k-latest": {
         "id": "ernie-4.0-8k-latest",
@@ -779,6 +863,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "ernie-4.0-8k": {
         "id": "ernie-4.0-8k",
@@ -786,6 +871,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "ernie-4.0-turbo-8k": {
         "id": "ernie-4.0-turbo-8k",
@@ -793,6 +879,7 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "ernie-4.5-8k-preview": {
         "id": "ernie-4.5-8k-preview",
@@ -800,47 +887,54 @@ ERNIE_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "max_output_tokens": 2048,
         "function_call_available": False,
         "response_format_available": True,
+        "native_multimodal": False,
     },
 }
 
 
 STEPFUN_DEFAULT_MODEL: Final[str] = "step-1-8k"
-STEPFUN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+STEPFUN_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "step-1-8k": {
         "id": "step-1-8k",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-1-32k": {
         "id": "step-1-32k",
         "context_length": 32000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-1-128k": {
         "id": "step-1-128k",
         "context_length": 128000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-1-256k": {
         "id": "step-1-256k",
         "context_length": 256000,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-2-16k": {
         "id": "step-2-16k",
         "context_length": 16384,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-1-flash": {
         "id": "step-1-flash",
         "context_length": 8192,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "step-1v-8k": {
         "id": "step-1v-8k",
@@ -860,18 +954,20 @@ STEPFUN_MODELS: Final[Dict[str, Dict[str, Any]]] = {
 
 
 XAI_DEFAULT_MODEL: Final[str] = "grok-2-latest"
-XAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
+XAI_MODELS: Final[Dict[str, ModelSettingDict]] = {
     "grok-beta": {
         "id": "grok-beta",
         "context_length": 131072,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "grok-2-latest": {
         "id": "grok-2-latest",
         "context_length": 131072,
         "function_call_available": True,
         "response_format_available": True,
+        "native_multimodal": False,
     },
     "grok-2-vision-latest": {
         "id": "grok-2-vision-latest",
@@ -880,4 +976,32 @@ XAI_MODELS: Final[Dict[str, Dict[str, Any]]] = {
         "response_format_available": True,
         "native_multimodal": True,
     },
+    "grok-3-beta": {
+        "id": "grok-beta",
+        "context_length": 131072,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": False,
+    },
+    "grok-3-fast-beta": {
+        "id": "grok-beta",
+        "context_length": 131072,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": False,
+    },
+    "grok-3-mini-beta": {
+        "id": "grok-beta",
+        "context_length": 131072,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": False,
+    },
+    "grok-3-mini-fast-beta": {
+        "id": "grok-beta",
+        "context_length": 131072,
+        "function_call_available": True,
+        "response_format_available": True,
+        "native_multimodal": False,
+    },
 }