feat(llm): add MiniMax as a built-in provider

Two regional endpoints (global api.minimax.io, China api.minimaxi.com)
with separate API keys. Models M2.7 / M2.5 plus -highspeed variants,
204K context. Follows the existing provider-preset pattern.

#789 #609 #577 #546 #395 #378

.env.example

@@ -6,4 +6,6 @@ XAI_API_KEY=
 DEEPSEEK_API_KEY=
 DASHSCOPE_API_KEY=
 ZHIPU_API_KEY=
+MINIMAX_API_KEY=
+MINIMAX_CN_API_KEY=
 OPENROUTER_API_KEY=

README.md

@@ -144,6 +144,8 @@ export XAI_API_KEY=...             # xAI (Grok)
 export DEEPSEEK_API_KEY=...        # DeepSeek
 export DASHSCOPE_API_KEY=...       # Qwen (Alibaba DashScope)
 export ZHIPU_API_KEY=...           # GLM (Zhipu)
+export MINIMAX_API_KEY=...         # MiniMax (global, api.minimax.io)
+export MINIMAX_CN_API_KEY=...      # MiniMax (China, api.minimaxi.com)
 export OPENROUTER_API_KEY=...      # OpenRouter
 export ALPHA_VANTAGE_API_KEY=...   # Alpha Vantage
 ```
@@ -184,7 +186,7 @@ An interface will appear showing results as they load, letting you track the age
 
 ### Implementation Details
 
-We built TradingAgents with LangGraph to ensure flexibility and modularity. The framework supports multiple LLM providers: OpenAI, Google, Anthropic, xAI, DeepSeek, Qwen (Alibaba DashScope), GLM (Zhipu), OpenRouter, Ollama for local models, and Azure OpenAI for enterprise.
+We built TradingAgents with LangGraph to ensure flexibility and modularity. The framework supports multiple LLM providers: OpenAI, Google, Anthropic, xAI, DeepSeek, Qwen (Alibaba DashScope), GLM (Zhipu), MiniMax (global + China), OpenRouter, Ollama for local models, and Azure OpenAI for enterprise.
 
 ### Python Usage
 
@@ -208,7 +210,7 @@ from tradingagents.graph.trading_graph import TradingAgentsGraph
 from tradingagents.default_config import DEFAULT_CONFIG
 
 config = DEFAULT_CONFIG.copy()
-config["llm_provider"] = "openai"        # openai, google, anthropic, xai, deepseek, qwen, glm, openrouter, ollama, azure
+config["llm_provider"] = "openai"        # openai, google, anthropic, xai, deepseek, qwen, glm, minimax, minimax-cn, openrouter, ollama, azure
 config["deep_think_llm"] = "gpt-5.4"     # Model for complex reasoning
 config["quick_think_llm"] = "gpt-5.4-mini" # Model for quick tasks
 config["max_debate_rounds"] = 2

cli/utils.py

@@ -239,6 +239,8 @@ def select_llm_provider() -> tuple[str, str | None]:
         ("DeepSeek", "deepseek", "https://api.deepseek.com"),
         ("Qwen", "qwen", "https://dashscope.aliyuncs.com/compatible-mode/v1"),
         ("GLM", "glm", "https://open.bigmodel.cn/api/paas/v4/"),
+        ("MiniMax", "minimax", "https://api.minimax.io/v1"),
+        ("MiniMax CN", "minimax-cn", "https://api.minimaxi.com/v1"),
         ("OpenRouter", "openrouter", "https://openrouter.ai/api/v1"),
         ("Azure OpenAI", "azure", None),
         ("Ollama", "ollama", "http://localhost:11434/v1"),

tradingagents/llm_clients/factory.py

@@ -4,7 +4,9 @@ from .base_client import BaseLLMClient
 
 # Providers that use the OpenAI-compatible chat completions API
 _OPENAI_COMPATIBLE = (
-    "openai", "xai", "deepseek", "qwen", "glm", "ollama", "openrouter",
+    "openai", "xai", "deepseek", "qwen", "glm",
+    "minimax", "minimax-cn",
+    "ollama", "openrouter",
 )
 
 

tradingagents/llm_clients/model_catalog.py

@@ -8,6 +8,22 @@ ModelOption = Tuple[str, str]
 ProviderModeOptions = Dict[str, Dict[str, List[ModelOption]]]
 
 
+# Shared model list for MiniMax's global and CN endpoints (same model IDs).
+_MINIMAX_MODELS: Dict[str, List[ModelOption]] = {
+    "quick": [
+        ("MiniMax M2.7 Highspeed — Fast, 204K ctx", "MiniMax-M2.7-highspeed"),
+        ("MiniMax M2.5 Highspeed — Previous-gen fast", "MiniMax-M2.5-highspeed"),
+        ("Custom model ID", "custom"),
+    ],
+    "deep": [
+        ("MiniMax M2.7 — Flagship, 204K ctx", "MiniMax-M2.7"),
+        ("MiniMax M2.5 — Previous-gen flagship", "MiniMax-M2.5"),
+        ("MiniMax M2.7 Highspeed — Faster M2.7, 204K ctx", "MiniMax-M2.7-highspeed"),
+        ("Custom model ID", "custom"),
+    ],
+}
+
+
 MODEL_OPTIONS: ProviderModeOptions = {
     "openai": {
         "quick": [
@@ -101,6 +117,10 @@ MODEL_OPTIONS: ProviderModeOptions = {
             ("Custom model ID", "custom"),
         ],
     },
+    # MiniMax: same model IDs across global (.io) and China (.com) regions,
+    # so the two provider keys share one model list.
+    "minimax": _MINIMAX_MODELS,
+    "minimax-cn": _MINIMAX_MODELS,
     # OpenRouter: fetched dynamically. Azure: any deployed model name.
     "ollama": {
         "quick": [

tradingagents/llm_clients/openai_client.py

@@ -119,6 +119,10 @@ _PROVIDER_CONFIG = {
     "deepseek": ("https://api.deepseek.com", "DEEPSEEK_API_KEY"),
     "qwen": ("https://dashscope-intl.aliyuncs.com/compatible-mode/v1", "DASHSCOPE_API_KEY"),
     "glm": ("https://api.z.ai/api/paas/v4/", "ZHIPU_API_KEY"),
+    # MiniMax exposes two regional endpoints with separate keys; mainland
+    # Chinese users hit .com while global users hit .io.
+    "minimax": ("https://api.minimax.io/v1", "MINIMAX_API_KEY"),
+    "minimax-cn": ("https://api.minimaxi.com/v1", "MINIMAX_CN_API_KEY"),
     "openrouter": ("https://openrouter.ai/api/v1", "OPENROUTER_API_KEY"),
     "ollama": ("http://localhost:11434/v1", None),
 }