cct-cli 0.7.9.0__py3-none-any.whl

calc_terminal/providers/adapters/anthropic_adapter.py

@@ -0,0 +1,127 @@
+"""
+calc_terminal/providers/adapters/anthropic_adapter.py
+====================================================
+Anthropic Claude Provider Adapter with live discovery and Claude 4 support.
+"""
+
+import requests
+from typing import List
+
+from .base import BaseProviderAdapter
+from ...models.schema import ModelInfo
+from ...models.validator import ModelValidator
+
+
+class AnthropicAdapter(BaseProviderAdapter):
+    provider_id = "anthropic"
+    display_name = "Anthropic (Claude)"
+    company = "Anthropic"
+    region = "US"
+    api_base = "https://api.anthropic.com/v1"
+    authentication_type = "header"
+    documentation_url = "https://docs.anthropic.com/en/docs/about-claude/models"
+
+    CATALOG_MODELS = {
+        "claude-4-sonnet-20260515": {
+            "display_name": "Claude 4 Sonnet",
+            "family": "Claude",
+            "version": "4",
+            "context_window": 500000,
+            "max_output_tokens": 32768,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 3.0, "output_price_per_1m": 15.0},
+            "recommendation_badges": ["Best for Coding"],
+        },
+        "claude-4-fable-5.1": {
+            "display_name": "Claude 4 Fable 5.1",
+            "family": "Claude",
+            "version": "4",
+            "context_window": 500000,
+            "max_output_tokens": 64000,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming", "long_context"],
+            "pricing": {"input_price_per_1m": 4.0, "output_price_per_1m": 16.0},
+            "recommendation_badges": ["Best for Research"],
+        },
+        "claude-sonnet-4": {
+            "display_name": "Claude Sonnet 4",
+            "family": "Claude",
+            "version": "4",
+            "context_window": 200000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 3.0, "output_price_per_1m": 15.0},
+            "recommendation_badges": ["Best for Coding"],
+        },
+        "claude-opus-4": {
+            "display_name": "Claude Opus 4",
+            "family": "Claude",
+            "version": "4",
+            "context_window": 200000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "vision", "research", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 15.0, "output_price_per_1m": 75.0},
+            "recommendation_badges": ["Best for Reasoning"],
+        },
+        "claude-3-7-sonnet": {
+            "display_name": "Claude 3.7 Sonnet",
+            "family": "Claude",
+            "version": "3.7",
+            "context_window": 200000,
+            "max_output_tokens": 8192,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 3.0, "output_price_per_1m": 15.0},
+            "recommendation_badges": ["Best for Coding"],
+        },
+        "claude-3-5-sonnet": {
+            "display_name": "Claude 3.5 Sonnet",
+            "family": "Claude",
+            "version": "3.5",
+            "context_window": 200000,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "vision", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 3.0, "output_price_per_1m": 15.0},
+        },
+        "claude-3-5-haiku": {
+            "display_name": "Claude 3.5 Haiku",
+            "family": "Claude",
+            "version": "3.5",
+            "context_window": 200000,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming", "fast"],
+            "pricing": {"input_price_per_1m": 0.8, "output_price_per_1m": 4.0},
+            "recommendation_badges": ["Best Fast Model"],
+        },
+    }
+
+    def discover_models(self, timeout: float = 8.0) -> List[ModelInfo]:
+        base = self.get_api_base()
+        key = self.get_api_key()
+        headers = {"Content-Type": "application/json", "anthropic-version": "2023-06-01"}
+        if key:
+            headers["x-api-key"] = key
+
+        discovered_ids = []
+        try:
+            resp = requests.get(f"{base}/models", headers=headers, timeout=timeout)
+            if resp.status_code == 200:
+                data = resp.json().get("data", [])
+                for item in data:
+                    mid = str(item.get("id", "")).strip()
+                    if mid:
+                        discovered_ids.append(mid)
+        except Exception:
+            pass
+
+        for cat_id in self.CATALOG_MODELS:
+            if cat_id not in discovered_ids:
+                discovered_ids.append(cat_id)
+
+        results = []
+        for mid in discovered_ids:
+            meta = dict(self.CATALOG_MODELS.get(mid, {}))
+            meta["model_id"] = mid
+            meta["endpoint"] = base
+            val = ModelValidator.validate_raw(meta, self.provider_id, self.config)
+            if val.is_valid and val.model_info:
+                results.append(val.model_info)
+        return results

calc_terminal/providers/adapters/base.py

@@ -0,0 +1,106 @@
+"""
+calc_terminal/providers/adapters/base.py
+========================================
+Base Provider Adapter Interface for Dynamic Model Discovery.
+
+Every provider adapter implements this contract to allow CAT to automatically
+discover, validate, and query AI models directly from official provider APIs.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Dict, List, Optional, Tuple, Any
+
+from ...models.schema import ModelInfo, STATUS_ACTIVE, STATUS_UNAVAILABLE, AVAILABILITY_PAID_API
+from ...models.validator import ModelValidator
+
+
+class BaseProviderAdapter(ABC):
+    """Abstract interface for all CAT provider adapters."""
+
+    provider_id: str = ""
+    display_name: str = ""
+    company: str = ""
+    region: str = "US"
+    api_base: str = ""
+    authentication_type: str = "bearer_token"  # bearer_token | query_param | none | custom
+    documentation_url: str = ""
+
+    def __init__(self, provider_id_or_config: Any = None, config: Optional[Dict[str, Any]] = None):
+        if isinstance(provider_id_or_config, str):
+            self.provider_id = provider_id_or_config
+            self.config = dict(config or {})
+        elif isinstance(provider_id_or_config, dict):
+            self.config = dict(provider_id_or_config)
+        else:
+            self.config = dict(config or {})
+
+    def get_api_base(self) -> str:
+        return (
+            self.config.get("base_url")
+            or self.config.get("api_url")
+            or self.api_base
+        ).rstrip("/")
+
+    def get_api_key(self) -> Optional[str]:
+        return self.config.get("api_key")
+
+    @abstractmethod
+    def discover_models(self, timeout: float = 8.0) -> List[ModelInfo]:
+        """Fetch and normalize model listings from official provider endpoint."""
+        pass
+
+    def validate_model(self, model_id: str) -> Tuple[bool, str]:
+        """Check if model ID is recognized and supported by this adapter."""
+        if not model_id or not model_id.strip():
+            return False, "Empty model identifier."
+        return True, "Valid model ID."
+
+    def get_model_metadata(self, model_id: str) -> Optional[ModelInfo]:
+        """Return curated or default ModelInfo for a specific model ID."""
+        res = ModelValidator.validate_raw(model_id, self.provider_id, self.config)
+        return res.model_info if res.is_valid else None
+
+    def get_capabilities(self, model_id: str) -> List[str]:
+        meta = self.get_model_metadata(model_id)
+        return meta.capabilities if meta else ["chat", "streaming"]
+
+    def get_limits(self, model_id: str) -> Dict[str, int]:
+        meta = self.get_model_metadata(model_id)
+        if meta:
+            return {"context_window": meta.context_window, "max_output_tokens": meta.max_output_tokens}
+        return {"context_window": 128000, "max_output_tokens": 4096}
+
+    def get_pricing(self, model_id: str) -> Dict[str, float]:
+        meta = self.get_model_metadata(model_id)
+        return meta.pricing if meta else {"input_price_per_1m": 0.0, "output_price_per_1m": 0.0}
+
+    def get_availability(self, model_id: str) -> str:
+        meta = self.get_model_metadata(model_id)
+        return meta.availability if meta else AVAILABILITY_PAID_API
+
+    def health_check(self, timeout: float = 5.0) -> Dict[str, Any]:
+        """Ping provider endpoint to verify connectivity and status."""
+        import requests
+        base = self.get_api_base()
+        if not base:
+            return {"status": "unknown", "provider": self.provider_id, "message": "No endpoint configured"}
+        try:
+            # Most OpenAI-compatible endpoints respond to GET /models or /
+            headers = {}
+            key = self.get_api_key()
+            if key and self.authentication_type == "bearer_token":
+                headers["Authorization"] = f"Bearer {key}"
+            resp = requests.get(f"{base}/models", headers=headers, timeout=timeout)
+            if resp.status_code == 401:
+                return {"status": "online", "provider": self.provider_id, "auth_required": True, "code": 401}
+            if resp.status_code < 400:
+                return {"status": "online", "provider": self.provider_id, "code": resp.status_code}
+            if resp.status_code >= 500:
+                return {"status": "degraded", "provider": self.provider_id, "code": resp.status_code}
+            return {"status": "online", "provider": self.provider_id, "code": resp.status_code}
+        except requests.exceptions.Timeout:
+            return {"status": "degraded", "provider": self.provider_id, "error": "Timeout"}
+        except requests.exceptions.ConnectionError:
+            return {"status": "offline", "provider": self.provider_id, "error": "Connection error"}
+        except Exception as e:
+            return {"status": "unknown", "provider": self.provider_id, "error": str(e)}

calc_terminal/providers/adapters/chinese_adapters.py

@@ -0,0 +1,420 @@
+"""
+calc_terminal/providers/adapters/chinese_adapters.py
+===================================================
+First-Class Adapters for Major Chinese AI Providers:
+  - DeepSeek (V3, R1 Reasoner)
+  - Alibaba Cloud / Qwen (Max, Plus, Turbo, Coder, QwQ)
+  - Moonshot AI (Kimi K3, K2, moonshot-v1)
+  - Z.AI / Zhipu (GLM-5.1, GLM-5.2, GLM-5.3, GLM-4)
+  - MiniMax (H3, abab6.5s)
+  - Tencent (Hunyuan / Hy3)
+  - ByteDance (Doubao / Seed)
+  - Xiaomi (MiMo)
+  - Baidu (ERNIE 4.0 / Qianfan)
+"""
+
+import requests
+from typing import List, Dict, Any, Optional
+
+from .base import BaseProviderAdapter
+from ...models.schema import ModelInfo, AVAILABILITY_PAID_API, AVAILABILITY_OPEN_WEIGHT
+from ...models.validator import ModelValidator
+
+
+class BaseOpenAICompatibleChineseAdapter(BaseProviderAdapter):
+    """Reusable base for OpenAI-compatible Chinese AI endpoints."""
+
+    catalog: Dict[str, Dict[str, Any]] = {}
+
+    def discover_models(self, timeout: float = 8.0) -> List[ModelInfo]:
+        base = self.get_api_base()
+        key = self.get_api_key()
+        headers = {"Content-Type": "application/json"}
+        if key:
+            headers["Authorization"] = f"Bearer {key}"
+
+        discovered_ids = []
+        try:
+            resp = requests.get(f"{base}/models", headers=headers, timeout=timeout)
+            if resp.status_code == 200:
+                data = resp.json().get("data", [])
+                for item in data:
+                    mid = str(item.get("id", "")).strip()
+                    if mid:
+                        discovered_ids.append(mid)
+        except Exception:
+            pass
+
+        # Ensure catalog models are present
+        for cat_id in self.catalog:
+            if cat_id not in discovered_ids:
+                discovered_ids.append(cat_id)
+
+        results = []
+        for mid in discovered_ids:
+            meta = dict(self.catalog.get(mid, {}))
+            meta["model_id"] = mid
+            meta["endpoint"] = base
+            val = ModelValidator.validate_raw(meta, self.provider_id, self.config)
+            if val.is_valid and val.model_info:
+                results.append(val.model_info)
+        return results
+
+
+class DeepSeekAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "deepseek"
+    display_name = "DeepSeek"
+    company = "DeepSeek"
+    region = "CN"
+    api_base = "https://api.deepseek.com/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://platform.deepseek.com/api-docs"
+
+    catalog = {
+        "deepseek-chat": {
+            "display_name": "DeepSeek-V3",
+            "family": "DeepSeek",
+            "version": "V3",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "coding", "tools", "streaming", "fast"],
+            "pricing": {"input_price_per_1m": 0.14, "output_price_per_1m": 0.28},
+            "availability": AVAILABILITY_OPEN_WEIGHT,
+            "recommendation_badges": ["Best Budget Model", "Best Fast Model"],
+        },
+        "deepseek-reasoner": {
+            "display_name": "DeepSeek-R1 (Reasoner)",
+            "family": "DeepSeek",
+            "version": "R1",
+            "context_window": 128000,
+            "max_output_tokens": 32768,
+            "reasoning_levels": ["high"],
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.55, "output_price_per_1m": 2.19},
+            "availability": AVAILABILITY_OPEN_WEIGHT,
+            "recommendation_badges": ["Best for Reasoning", "Best Budget Model"],
+        },
+        "deepseek-r1": {
+            "display_name": "DeepSeek-R1",
+            "family": "DeepSeek",
+            "version": "R1",
+            "context_window": 128000,
+            "max_output_tokens": 32768,
+            "reasoning_levels": ["high"],
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.55, "output_price_per_1m": 2.19},
+            "availability": AVAILABILITY_OPEN_WEIGHT,
+            "recommendation_badges": ["Best for Reasoning", "Best Budget Model"],
+        },
+    }
+
+
+class QwenAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "alibaba"
+    display_name = "Alibaba Cloud (Qwen)"
+    company = "Alibaba Cloud"
+    region = "CN"
+    api_base = "https://dashscope.aliyuncs.com/compatible-mode/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://help.aliyun.com/zh/dashscope"
+
+    catalog = {
+        "qwen-max": {
+            "display_name": "Qwen Max",
+            "family": "Qwen",
+            "context_window": 131072,
+            "max_output_tokens": 8192,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 2.8, "output_price_per_1m": 8.4},
+            "recommendation_badges": ["Best for Coding"],
+        },
+        "qwen-plus": {
+            "display_name": "Qwen Plus",
+            "family": "Qwen",
+            "context_window": 131072,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.4, "output_price_per_1m": 1.2},
+        },
+        "qwen-turbo": {
+            "display_name": "Qwen Turbo",
+            "family": "Qwen",
+            "context_window": 1000000,
+            "max_output_tokens": 8192,
+            "capabilities": ["tools", "streaming", "fast", "long_context"],
+            "pricing": {"input_price_per_1m": 0.05, "output_price_per_1m": 0.15},
+            "recommendation_badges": ["Best Fast Model", "Best Budget Model"],
+        },
+        "qwen-2.5-coder-32b": {
+            "display_name": "Qwen 2.5 Coder 32B",
+            "family": "Qwen",
+            "context_window": 131072,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.5, "output_price_per_1m": 1.5},
+            "availability": AVAILABILITY_OPEN_WEIGHT,
+            "recommendation_badges": ["Best for Coding"],
+        },
+        "qwen-qwq-32b": {
+            "display_name": "Qwen QwQ 32B (Reasoning)",
+            "family": "Qwen",
+            "context_window": 131072,
+            "max_output_tokens": 32768,
+            "reasoning_levels": ["high"],
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.8, "output_price_per_1m": 2.4},
+            "availability": AVAILABILITY_OPEN_WEIGHT,
+            "recommendation_badges": ["Best for Reasoning"],
+        },
+    }
+
+
+class MoonshotAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "moonshot"
+    display_name = "Moonshot AI (Kimi)"
+    company = "Moonshot AI"
+    region = "CN"
+    api_base = "https://api.moonshot.cn/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://platform.moonshot.cn/docs"
+
+    catalog = {
+        "kimi-k3": {
+            "display_name": "Kimi K3",
+            "family": "Kimi",
+            "context_window": 256000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "long_context", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.5, "output_price_per_1m": 6.0},
+            "recommendation_badges": ["Best for Research"],
+        },
+        "kimi-k2": {
+            "display_name": "Kimi K2",
+            "family": "Kimi",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.0, "output_price_per_1m": 4.0},
+        },
+        "moonshot-v1-128k": {
+            "display_name": "Moonshot v1 128k",
+            "family": "Kimi",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["long_context", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.2, "output_price_per_1m": 4.8},
+        },
+    }
+
+
+class ZhipuGLMAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "zhipu"
+    display_name = "Zhipu AI (GLM)"
+    company = "Zhipu AI / Z.AI"
+    region = "CN"
+    api_base = "https://open.bigmodel.cn/api/paas/v4"
+    authentication_type = "bearer_token"
+    documentation_url = "https://open.bigmodel.cn/dev/api"
+
+    catalog = {
+        "glm-5.1": {
+            "display_name": "GLM-5.1",
+            "family": "GLM",
+            "version": "5.1",
+            "context_window": 256000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.5, "output_price_per_1m": 5.0},
+            "recommendation_badges": ["Best for Reasoning"],
+        },
+        "glm-5.2": {
+            "display_name": "GLM-5.2",
+            "family": "GLM",
+            "version": "5.2",
+            "context_window": 256000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 2.0, "output_price_per_1m": 6.0},
+        },
+        "glm-5.3": {
+            "display_name": "GLM-5.3",
+            "family": "GLM",
+            "version": "5.3",
+            "context_window": 512000,
+            "max_output_tokens": 32768,
+            "capabilities": ["reasoning", "coding", "vision", "tools", "streaming", "long_context"],
+            "pricing": {"input_price_per_1m": 2.5, "output_price_per_1m": 8.0},
+            "recommendation_badges": ["Best for Coding", "Best for Research"],
+        },
+        "glm-4-plus": {
+            "display_name": "GLM-4 Plus",
+            "family": "GLM",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.4, "output_price_per_1m": 4.2},
+        },
+        "glm-4-air": {
+            "display_name": "GLM-4 Air",
+            "family": "GLM",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["tools", "streaming", "fast"],
+            "pricing": {"input_price_per_1m": 0.14, "output_price_per_1m": 0.42},
+            "recommendation_badges": ["Best Fast Model"],
+        },
+    }
+
+
+class MiniMaxAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "minimax"
+    display_name = "MiniMax"
+    company = "MiniMax"
+    region = "CN"
+    api_base = "https://api.minimax.chat/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://platform.minimaxi.com"
+
+    catalog = {
+        "h3": {
+            "display_name": "MiniMax H3",
+            "family": "MiniMax",
+            "context_window": 256000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.0, "output_price_per_1m": 4.0},
+        },
+        "minimax-text-01": {
+            "display_name": "MiniMax Text 01",
+            "family": "MiniMax",
+            "context_window": 1000000,
+            "max_output_tokens": 16384,
+            "capabilities": ["long_context", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.5, "output_price_per_1m": 2.0},
+        },
+        "abab6.5s": {
+            "display_name": "MiniMax Abab 6.5s",
+            "family": "MiniMax",
+            "context_window": 256000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "fast", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.2, "output_price_per_1m": 0.8},
+            "recommendation_badges": ["Best Fast Model"],
+        },
+    }
+
+
+class TencentHunyuanAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "tencent"
+    display_name = "Tencent Hunyuan"
+    company = "Tencent"
+    region = "CN"
+    api_base = "https://api.hunyuan.cloud.tencent.com/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://cloud.tencent.com/document/product/1729"
+
+    catalog = {
+        "hy3": {
+            "display_name": "Hunyuan Hy3",
+            "family": "Hunyuan",
+            "context_window": 256000,
+            "max_output_tokens": 16384,
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.5, "output_price_per_1m": 5.0},
+        },
+        "hunyuan-pro": {
+            "display_name": "Hunyuan Pro",
+            "family": "Hunyuan",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 1.0, "output_price_per_1m": 4.0},
+        },
+    }
+
+
+class ByteDanceSeedAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "bytedance"
+    display_name = "ByteDance (Doubao / Seed)"
+    company = "ByteDance"
+    region = "CN"
+    api_base = "https://ark.cn-beijing.volces.com/api/v3"
+    authentication_type = "bearer_token"
+    documentation_url = "https://www.volcengine.com/docs/82379"
+
+    catalog = {
+        "doubao-pro": {
+            "display_name": "Doubao Pro (Seed)",
+            "family": "Seed",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.12, "output_price_per_1m": 0.24},
+            "recommendation_badges": ["Best Budget Model"],
+        },
+        "doubao-lite": {
+            "display_name": "Doubao Lite (Seed)",
+            "family": "Seed",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "fast", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.04, "output_price_per_1m": 0.08},
+            "recommendation_badges": ["Best Fast Model"],
+        },
+    }
+
+
+class XiaomiMiMoAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "xiaomi"
+    display_name = "Xiaomi MiMo"
+    company = "Xiaomi"
+    region = "CN"
+    api_base = "https://api.mimo.mi.com/v1"
+    authentication_type = "bearer_token"
+    documentation_url = "https://dev.mi.com"
+
+    catalog = {
+        "mimo-pro": {
+            "display_name": "MiMo Pro",
+            "family": "MiMo",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 0.5, "output_price_per_1m": 1.5},
+        },
+    }
+
+
+class BaiduErnieAdapter(BaseOpenAICompatibleChineseAdapter):
+    provider_id = "baidu"
+    display_name = "Baidu (ERNIE)"
+    company = "Baidu"
+    region = "CN"
+    api_base = "https://aip.baidubce.com/v2"
+    authentication_type = "bearer_token"
+    documentation_url = "https://cloud.baidu.com/doc/WENXINWORKSHOP"
+
+    catalog = {
+        "ernie-4.0-turbo": {
+            "display_name": "ERNIE 4.0 Turbo",
+            "family": "ERNIE",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["reasoning", "coding", "tools", "streaming"],
+            "pricing": {"input_price_per_1m": 2.0, "output_price_per_1m": 6.0},
+        },
+        "ernie-speed": {
+            "display_name": "ERNIE Speed",
+            "family": "ERNIE",
+            "context_window": 128000,
+            "max_output_tokens": 8192,
+            "capabilities": ["chat", "fast", "streaming"],
+            "pricing": {"input_price_per_1m": 0.0, "output_price_per_1m": 0.0},
+            "recommendation_badges": ["Best Free Model"],
+        },
+    }
+
+
+# Convenience aliases
+AlibabaQwenAdapter = QwenAdapter
+MoonshotKimiAdapter = MoonshotAdapter