feat: implement zhipu and alibaba coding plan providers

frontend/src/components/pages/models.tsx

@@ -110,6 +110,26 @@ const LOCAL_PROVIDER_CONFIGS: Record<string, Partial<ProviderConfig>> = {
     category: ["llm", "embedding"],
     defaultBaseUrl: "https://api.openai.com/v1"
   },
+  "zai-coding-plan": {
+    icon: <img src="/zhipu.svg" alt="Z.AI" className="w-6 h-6" />,
+    category: ["llm"],
+    defaultBaseUrl: "https://api.z.ai/api/coding/paas/v4"
+  },
+  "zhipuai-coding-plan": {
+    icon: <img src="/zhipu.svg" alt="Zhipu" className="w-6 h-6" />,
+    category: ["llm"],
+    defaultBaseUrl: "https://open.bigmodel.cn/api/coding/paas/v4"
+  },
+  "alibaba-coding-plan": {
+    icon: <img src="/dashscope.png" alt="Alibaba Bailian" className="w-6 h-6" />,
+    category: ["llm"],
+    defaultBaseUrl: "https://coding-intl.dashscope.aliyuncs.com/v1"
+  },
+  "alibaba-coding-plan-cn": {
+    icon: <img src="/dashscope.png" alt="Alibaba Bailian" className="w-6 h-6" />,
+    category: ["llm"],
+    defaultBaseUrl: "https://coding.dashscope.aliyuncs.com/v1"
+  },
   azure_openai: {
     icon: <Zap className="w-6 h-6 text-blue-500" />,
     category: ["llm"]

frontend/src/i18n/locales/en.ts

@@ -1242,6 +1242,18 @@ Build when you need.`
       openai: {
         description: "Access GPT-4o, GPT-4 Turbo, and DALL-E 3 models. Industry standard for reasoning and creativity.",
       },
+      "zai-coding-plan": {
+        description: "GLM Coding Plan endpoints via Z.AI.",
+      },
+      "zhipuai-coding-plan": {
+        description: "GLM Coding Plan endpoints via Zhipu AI.",
+      },
+      "alibaba-coding-plan": {
+        description: "Alibaba Bailian Coding Plan endpoint.",
+      },
+      "alibaba-coding-plan-cn": {
+        description: "Alibaba Bailian Coding Plan endpoint (China).",
+      },
       azure_openai: {
         description: "Enterprise-grade OpenAI models hosted on Azure.",
       },

frontend/src/i18n/locales/zh.ts

@@ -1242,6 +1242,18 @@ Build when you need.`
       openai: {
         description: "访问 GPT-4o、GPT-4 Turbo 和 DALL-E 3 模型。推理和创造力的行业标准。",
       },
+      "zai-coding-plan": {
+        description: "Z.AI GLM Coding Plan 端点。",
+      },
+      "zhipuai-coding-plan": {
+        description: "智谱 GLM Coding Plan 端点（中国区）。",
+      },
+      "alibaba-coding-plan": {
+        description: "阿里云百炼 Coding Plan 端点。",
+      },
+      "alibaba-coding-plan-cn": {
+        description: "阿里云百炼 Coding Plan 端点（中国区）。",
+      },
       azure_openai: {
         description: "Azure 托管的企业级 OpenAI 模型。",
       },

src/xagent/core/model/chat/basic/adapter.py

@@ -1,6 +1,10 @@
 import os
 
 from ....model import ChatModelConfig, ModelConfig
+from ....model.providers import (
+    default_base_url_for_provider,
+    provider_compatibility_for_provider,
+)
 from ....retry import create_retry_wrapper
 from ..error import retry_on
 from .azure_openai import AzureOpenAILLM
@@ -19,11 +23,25 @@ def create_base_llm(model: ModelConfig) -> BaseLLM:
     if not isinstance(model, ChatModelConfig):
         raise TypeError(f"Invalid model type: {type(model).__name__}")
 
-    if model.model_provider == "openai":
+    compatibility = provider_compatibility_for_provider(model.model_provider)
+
+    if compatibility == "openai_compatible":
         llm: BaseLLM = OpenAILLM(
             model_name=model.model_name,
             api_key=model.api_key,
-            base_url=model.base_url,
+            base_url=model.base_url
+            or default_base_url_for_provider(model.model_provider),
+            default_temperature=model.default_temperature,
+            default_max_tokens=model.default_max_tokens,
+            timeout=model.timeout,
+            abilities=model.abilities,
+        )
+    elif compatibility == "claude_compatible":
+        llm = ClaudeLLM(
+            model_name=model.model_name,
+            api_key=model.api_key,
+            base_url=model.base_url
+            or default_base_url_for_provider(model.model_provider),
             default_temperature=model.default_temperature,
             default_max_tokens=model.default_max_tokens,
             timeout=model.timeout,
@@ -60,16 +78,6 @@ def create_base_llm(model: ModelConfig) -> BaseLLM:
             timeout=model.timeout,
             abilities=model.abilities,
         )
-    elif model.model_provider == "claude":
-        llm = ClaudeLLM(
-            model_name=model.model_name,
-            api_key=model.api_key,
-            base_url=model.base_url,
-            default_temperature=model.default_temperature,
-            default_max_tokens=model.default_max_tokens,
-            timeout=model.timeout,
-            abilities=model.abilities,
-        )
     elif model.model_provider == "xinference":
         llm = XinferenceLLM(
             model_name=model.model_name,

src/xagent/core/model/chat/langchain.py

@@ -10,6 +10,10 @@
 from langchain_openai import AzureChatOpenAI, ChatOpenAI
 
 from ...model import ChatModelConfig, ModelConfig
+from ...model.providers import (
+    default_base_url_for_provider,
+    provider_compatibility_for_provider,
+)
 from ...retry import ExponentialBackoff, RetryStrategy, create_retry_wrapper
 from .error import retry_on
 
@@ -97,14 +101,16 @@ def create_base_chat_model(
         raise TypeError(f"Unsupported Chat model type: {type(model).__name__}")
 
     temp = temperature if temperature is not None else model.default_temperature
+    compatibility = provider_compatibility_for_provider(model.model_provider)
 
-    if model.model_provider == "openai":
+    if compatibility == "openai_compatible":
         return ChatOpenAI(
             model=model.model_name,
             temperature=temp,
             max_tokens=model.default_max_tokens,
             api_key=model.api_key,
-            base_url=model.base_url,
+            base_url=model.base_url
+            or default_base_url_for_provider(model.model_provider),
             timeout=model.timeout,
         )
     elif model.model_provider == "zhipu":

src/xagent/core/model/providers.py

@@ -0,0 +1,146 @@
+from typing import Any, Optional
+
+_PROVIDER_ALIASES: dict[str, str] = {
+    "zai_coding_plan": "zai-coding-plan",
+    "zhipuai_coding_plan": "zhipuai-coding-plan",
+    "alibaba_coding_plan": "alibaba-coding-plan",
+    "alibaba_coding_plan_cn": "alibaba-coding-plan-cn",
+}
+
+# Provider default base URLs used when callers omit an explicit base URL.
+_DEFAULT_BASE_URL_BY_PROVIDER: dict[str, str] = {
+    "openai": "https://api.openai.com/v1",
+    "dashscope": "https://dashscope.aliyuncs.com/compatible-mode/v1",
+    "zhipu": "https://open.bigmodel.cn/api/paas/v4",
+    # Opencode / models.dev naming
+    "zai-coding-plan": "https://api.z.ai/api/coding/paas/v4",
+    "zhipuai-coding-plan": "https://open.bigmodel.cn/api/coding/paas/v4",
+    # Alibaba Bailian (Model Studio) coding plan
+    "alibaba-coding-plan": "https://coding-intl.dashscope.aliyuncs.com/v1",
+    "alibaba-coding-plan-cn": "https://coding.dashscope.aliyuncs.com/v1",
+}
+
+_CURATED_MODELS_BY_PROVIDER: dict[str, tuple[str, ...]] = {
+    "alibaba-coding-plan": (
+        "glm-4.7",
+        "glm-5",
+        "qwen3-coder-next",
+        "qwen3-coder-plus",
+        "qwen3-max-2026-01-23",
+        "qwen3.5-plus",
+    ),
+    "alibaba-coding-plan-cn": (
+        "glm-4.7",
+        "glm-5",
+        "qwen3-coder-next",
+        "qwen3-coder-plus",
+        "qwen3-max-2026-01-23",
+        "qwen3.5-plus",
+    ),
+}
+
+_SUPPORTED_PROVIDER_METADATA: tuple[dict[str, Any], ...] = (
+    {
+        "id": "openai",
+        "name": "OpenAI",
+        "description": "OpenAI API compatible models",
+        "requires_base_url": False,
+        "compatibility": "openai_compatible",
+    },
+    {
+        "id": "claude",
+        "name": "Anthropic Claude",
+        "description": "Anthropic's Claude models",
+        "requires_base_url": False,
+        "compatibility": "claude_compatible",
+    },
+    {
+        "id": "gemini",
+        "name": "Google Gemini",
+        "description": "Google's Gemini models",
+        "requires_base_url": False,
+    },
+    {
+        "id": "xinference",
+        "name": "Xinference",
+        "description": "Xinference models for local inference",
+        "requires_base_url": True,
+    },
+    {
+        "id": "dashscope",
+        "name": "DashScope",
+        "description": "Alibaba Cloud's DashScope models",
+        "requires_base_url": False,
+    },
+    {
+        "id": "alibaba-coding-plan",
+        "name": "Alibaba Coding Plan",
+        "description": "Alibaba Bailian (Model Studio) coding plan",
+        "requires_base_url": False,
+        "compatibility": "openai_compatible",
+    },
+    {
+        "id": "alibaba-coding-plan-cn",
+        "name": "Alibaba Coding Plan (China)",
+        "description": "Alibaba Bailian (Model Studio) coding plan (China)",
+        "requires_base_url": False,
+        "compatibility": "openai_compatible",
+    },
+    {
+        "id": "zhipu",
+        "name": "Zhipu AI",
+        "description": "Zhipu AI models (GLM series) using zai SDK",
+        "requires_base_url": False,
+    },
+    {
+        "id": "zai-coding-plan",
+        "name": "Z.AI Coding Plan",
+        "description": "GLM coding plan via Z.AI",
+        "requires_base_url": False,
+        "compatibility": "openai_compatible",
+    },
+    {
+        "id": "zhipuai-coding-plan",
+        "name": "Zhipu AI Coding Plan",
+        "description": "GLM coding plan via Zhipu AI",
+        "requires_base_url": False,
+        "compatibility": "openai_compatible",
+    },
+)
+
+
+def _normalize_provider(provider: str) -> str:
+    return provider.lower().strip()
+
+
+def canonical_provider_name(provider: str) -> str:
+    normalized = _normalize_provider(provider)
+    return _PROVIDER_ALIASES.get(normalized, normalized)
+
+
+def default_base_url_for_provider(provider: str) -> Optional[str]:
+    return _DEFAULT_BASE_URL_BY_PROVIDER.get(canonical_provider_name(provider))
+
+
+def curated_models_for_provider(provider: str) -> tuple[str, ...]:
+    return _CURATED_MODELS_BY_PROVIDER.get(canonical_provider_name(provider), ())
+
+
+def provider_compatibility_for_provider(provider: str) -> Optional[str]:
+    provider_id = canonical_provider_name(provider)
+    for provider_info in _SUPPORTED_PROVIDER_METADATA:
+        if provider_info["id"] == provider_id:
+            compatibility = provider_info.get("compatibility")
+            return str(compatibility) if compatibility is not None else None
+    return None
+
+
+def get_supported_provider_metadata() -> list[dict[str, Any]]:
+    providers: list[dict[str, Any]] = []
+    for provider in _SUPPORTED_PROVIDER_METADATA:
+        provider_info = dict(provider)
+        default_base_url = default_base_url_for_provider(provider_info["id"])
+        if default_base_url is not None:
+            provider_info["default_base_url"] = default_base_url
+        providers.append(provider_info)
+    return providers

src/xagent/web/api/model.py

@@ -13,6 +13,7 @@
     ImageModelConfig,
     ModelConfig,
 )
+from xagent.core.model.providers import default_base_url_for_provider
 from xagent.core.utils.security import redact_sensitive_text
 
 from ..auth_dependencies import get_current_user
@@ -59,12 +60,14 @@ async def create_model(
             detail="Only administrators can share models with all users",
         )
 
+    base_url = model.base_url or default_base_url_for_provider(model.model_provider)
+
     if model.category == "llm":
         config: ModelConfig = ChatModelConfig(
             id=model.model_id,
             model_name=model.model_name,
             model_provider=model.model_provider,
-            base_url=model.base_url,
+            base_url=base_url,
             api_key=model.api_key,
             default_temperature=model.temperature,
             timeout=180.0,
@@ -76,7 +79,7 @@ async def create_model(
             id=model.model_id,
             model_name=model.model_name,
             model_provider=model.model_provider,
-            base_url=model.base_url,
+            base_url=base_url,
             api_key=model.api_key,
             timeout=180.0,
             abilities=model.abilities,
@@ -88,7 +91,7 @@ async def create_model(
             id=model.model_id,
             model_name=model.model_name,
             model_provider=model.model_provider,
-            base_url=model.base_url,
+            base_url=base_url,
             api_key=model.api_key,
             default_temperature=model.temperature,
             timeout=180.0,

src/xagent/web/schemas/model.py

@@ -334,6 +334,8 @@ class ProviderInfo(BaseModel):
     name: str
     description: str
     requires_base_url: bool
+    default_base_url: Optional[str] = None
+    compatibility: Optional[str] = None
 
 
 class SupportedProvidersResponse(BaseModel):