NousResearch · cudasuan · May 30, 2026
diff --git a/agent/model_metadata.py b/agent/model_metadata.py
@@ -47,7 +47,7 @@ def _resolve_requests_verify() -> bool | str:
 _PROVIDER_PREFIXES: frozenset[str] = frozenset({
     "openrouter", "nous", "openai-codex", "copilot", "copilot-acp",
     "gemini", "ollama-cloud", "zai", "kimi-coding", "kimi-coding-cn", "stepfun", "minimax", "minimax-oauth", "minimax-cn", "anthropic", "deepseek",
-    "opencode-zen", "opencode-go", "kilocode", "alibaba", "novita",
+    "opencode-zen", "opencode-go", "kilocode", "alibaba", "alibaba-token-plan", "novita",
     "qwen-oauth",
     "xiaomi",
     "arcee",
@@ -354,6 +354,7 @@ def _is_custom_endpoint(base_url: str) -> bool:
     "api.minimax": "minimax",
     "dashscope.aliyuncs.com": "alibaba",
     "dashscope-intl.aliyuncs.com": "alibaba",
+    "token-plan.cn-beijing.maas.aliyuncs.com": "alibaba-token-plan",
     "portal.qwen.ai": "qwen-oauth",
     "openrouter.ai": "openrouter",
     "generativelanguage.googleapis.com": "gemini",

diff --git a/agent/system_prompt.py b/agent/system_prompt.py
@@ -194,12 +194,13 @@ def build_system_prompt_parts(agent: Any, system_message: Optional[str] = None)
     if skills_prompt:
         stable_parts.append(skills_prompt)
 
-    # Alibaba Coding Plan API always returns "glm-4.7" as model name regardless
-    # of the requested model. Inject explicit model identity into the system prompt
-    # so the agent can correctly report which model it is (workaround for API bug).
+    # Alibaba DashScope / Token Plan APIs may return a different model name than
+    # requested (e.g. returning "glm-4.7" regardless of the actual model).
+    # Inject explicit model identity into the system prompt so the agent can
+    # correctly report which model it is (workaround for API bug).
     # Stable for the lifetime of an agent instance — model and provider are fixed
     # at construction time.
-    if agent.provider == "alibaba":
+    if agent.provider in ("alibaba", "alibaba-coding-plan", "alibaba-token-plan"):
         _model_short = agent.model.split("/")[-1] if "/" in agent.model else agent.model
         stable_parts.append(
             f"You are powered by the model named {_model_short}. "

diff --git a/hermes_cli/auth.py b/hermes_cli/auth.py
@@ -330,6 +330,14 @@ class ProviderConfig:
         api_key_env_vars=("ALIBABA_CODING_PLAN_API_KEY", "DASHSCOPE_API_KEY"),
         base_url_env_var="ALIBABA_CODING_PLAN_BASE_URL",
     ),
+    "alibaba-token-plan": ProviderConfig(
+        id="alibaba-token-plan",
+        name="Alibaba Cloud (Token Plan)",
+        auth_type="api_key",
+        inference_base_url="https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
+        api_key_env_vars=("ALIBABA_TOKEN_PLAN_API_KEY", "DASHSCOPE_API_KEY"),
+        base_url_env_var="ALIBABA_TOKEN_PLAN_BASE_URL",
+    ),
     "minimax-cn": ProviderConfig(
         id="minimax-cn",
         name="MiniMax (China)",
@@ -1495,6 +1503,8 @@ def resolve_provider(
         "minimax-portal": "minimax-oauth", "minimax-global": "minimax-oauth", "minimax_oauth": "minimax-oauth",
         "alibaba_coding": "alibaba-coding-plan", "alibaba-coding": "alibaba-coding-plan",
         "alibaba_coding_plan": "alibaba-coding-plan",
+        "alibaba_token_plan": "alibaba-token-plan", "alibaba-token": "alibaba-token-plan",
+        "aliyun-token-plan": "alibaba-token-plan", "token-plan": "alibaba-token-plan",
         "claude": "anthropic", "claude-code": "anthropic",
         "github": "copilot", "github-copilot": "copilot",
         "github-models": "copilot", "github-model": "copilot",

diff --git a/hermes_cli/doctor.py b/hermes_cli/doctor.py
@@ -371,6 +371,7 @@ def _build_apikey_providers_list() -> list:
         ("Hugging Face",     ("HF_TOKEN",),                                  "https://router.huggingface.co/v1/models", "HF_BASE_URL", True),
         ("NVIDIA NIM",       ("NVIDIA_API_KEY",),                            "https://integrate.api.nvidia.com/v1/models", "NVIDIA_BASE_URL", True),
         ("Alibaba/DashScope", ("DASHSCOPE_API_KEY",),                        "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/models", "DASHSCOPE_BASE_URL", True),
+        ("Alibaba Token Plan", ("ALIBABA_TOKEN_PLAN_API_KEY", "DASHSCOPE_API_KEY"), "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1/models", "ALIBABA_TOKEN_PLAN_BASE_URL", True),
         # MiniMax global: /v1 endpoint supports /models.
         ("MiniMax",          ("MINIMAX_API_KEY",),                           "https://api.minimax.io/v1/models",    "MINIMAX_BASE_URL", True),
         # MiniMax CN: /v1 endpoint does NOT support /models (returns 404).
@@ -389,7 +390,8 @@ def _build_apikey_providers_list() -> list:
         "StepFun Step Plan": "stepfun", "Kimi / Moonshot (China)": "kimi-coding-cn",
         "Arcee AI": "arcee", "GMI Cloud": "gmi", "DeepSeek": "deepseek",
         "Hugging Face": "huggingface", "NVIDIA NIM": "nvidia",
-        "Alibaba/DashScope": "alibaba", "MiniMax": "minimax",
+        "Alibaba/DashScope": "alibaba", "Alibaba Token Plan": "alibaba-token-plan",
+        "MiniMax": "minimax",
         "MiniMax (China)": "minimax-cn",
         "Kilo Code": "kilocode", "OpenCode Zen": "opencode-zen",
         "OpenCode Go": "opencode-go",

diff --git a/hermes_cli/models.py b/hermes_cli/models.py
@@ -421,6 +421,22 @@ def _xai_curated_models() -> list[str]:
         "glm-4.7",
         "MiniMax-M2.5",
     ],
+    # Alibaba Token Plan (Team Edition) — subscription-based Credits billing,
+    # endpoint: token-plan.cn-beijing.maas.aliyuncs.com.
+    # Docs: https://help.aliyun.com/zh/model-studio/token-plan-overview
+    "alibaba-token-plan": [
+        "qwen3.7-max",
+        "qwen3.6-plus",
+        "qwen3.6-flash",
+        "deepseek-v4-pro",
+        "deepseek-v4-flash",
+        "deepseek-v3.2",
+        "kimi-k2.6",
+        "kimi-k2.5",
+        "glm-5.1",
+        "glm-5",
+        "MiniMax-M2.5",
+    ],
     # Curated HF model list — only agentic models that map to OpenRouter defaults.
     "huggingface": [
         "moonshotai/Kimi-K2.5",

diff --git a/hermes_cli/providers.py b/hermes_cli/providers.py
@@ -143,6 +143,10 @@ class HermesOverlay:
         transport="openai_chat",
         base_url_env_var="ALIBABA_CODING_PLAN_BASE_URL",
     ),
+    "alibaba-token-plan": HermesOverlay(
+        transport="openai_chat",
+        base_url_env_var="ALIBABA_TOKEN_PLAN_BASE_URL",
+    ),
     "opencode": HermesOverlay(
         transport="openai_chat",
         is_aggregator=True,
@@ -311,6 +315,12 @@ class ProviderDef:
     "alibaba-coding": "alibaba-coding-plan",
     "alibaba_coding_plan": "alibaba-coding-plan",
 
+    # alibaba-token-plan
+    "alibaba_token_plan": "alibaba-token-plan",
+    "alibaba-token": "alibaba-token-plan",
+    "aliyun-token-plan": "alibaba-token-plan",
+    "token-plan": "alibaba-token-plan",
+
     # google-gemini-cli (OAuth + Code Assist)
     "gemini-cli": "google-gemini-cli",
     "gemini-oauth": "google-gemini-cli",

diff --git a/plugins/model-providers/alibaba-token-plan/__init__.py b/plugins/model-providers/alibaba-token-plan/__init__.py
@@ -0,0 +1,24 @@
+"""Alibaba Cloud Token Plan (Team Edition) provider profile.
+
+Separate from the standard `alibaba` and `alibaba-coding-plan` profiles because
+it hits a different endpoint (token-plan.cn-beijing.maas.aliyuncs.com) with
+its own subscription-based billing (Credits).
+
+Docs: https://help.aliyun.com/zh/model-studio/token-plan-overview
+"""
+
+from providers import register_provider
+from providers.base import ProviderProfile
+
+alibaba_token_plan = ProviderProfile(
+    name="alibaba-token-plan",
+    aliases=("alibaba_token_plan", "aliyun-token-plan", "token-plan"),
+    display_name="Alibaba Cloud (Token Plan)",
+    description="Alibaba Cloud Token Plan — team subscription (Credits)",
+    signup_url="https://www.aliyun.com/benefit/scene/tokenplan",
+    env_vars=("ALIBABA_TOKEN_PLAN_API_KEY", "DASHSCOPE_API_KEY", "ALIBABA_TOKEN_PLAN_BASE_URL"),
+    base_url="https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
+    auth_type="api_key",
+)
+
+register_provider(alibaba_token_plan)
diff --git a/plugins/model-providers/alibaba-token-plan/plugin.yaml b/plugins/model-providers/alibaba-token-plan/plugin.yaml
@@ -0,0 +1,5 @@
+name: alibaba-token-plan-provider
+kind: model-provider
+version: 1.0.0
+description: Alibaba Cloud Token Plan (Team Edition)
+author: Nous Research
diff --git a/tests/providers/test_plugin_discovery.py b/tests/providers/test_plugin_discovery.py
@@ -68,7 +68,7 @@ def test_all_profiles_register():
     # Spot-check representative providers from different categories
     for required in (
         "openrouter", "anthropic", "custom", "bedrock", "openai-codex",
-        "minimax-oauth", "gmi", "xiaomi", "alibaba-coding-plan",
+        "minimax-oauth", "gmi", "xiaomi", "alibaba-coding-plan", "alibaba-token-plan",
     ):
         assert required in names, f"Missing profile: {required}"