diff --git a/.env.example b/.env.example
index 351ad4e..761e195 100644
--- a/.env.example
+++ b/.env.example
@@ -4,6 +4,10 @@ ANTHROPIC_API_KEY=
 OPENAI_API_KEY=
 
 # chat-latest follows ChatGPT Instant. gpt-* uses OpenAI; claude-* uses Anthropic.
+# vendor/model names (e.g. deepseek/deepseek-chat) use an OpenAI-compatible
+# gateway: set OPENAI_BASE_URL (e.g. https://openrouter.ai/api/v1) and put the
+# gateway key in OPENAI_API_KEY. Server keys only; browser keys use OpenAI.
+# OPENAI_BASE_URL=
 # Model access/availability must be verified with your provider.
 # Blank or unsupported model prefixes fail at startup.
 # Keys are optional for BYOK-only hosting; visitors enter both keys in the app.
diff --git a/apps/agent/src/model.py b/apps/agent/src/model.py
index 33725ab..38d0a65 100644
--- a/apps/agent/src/model.py
+++ b/apps/agent/src/model.py
@@ -21,11 +21,18 @@
 
 def build_model(*, allow_unconfigured: bool = False) -> BaseChatModel:
     model_name = os.environ.get("LLM_MODEL", DEFAULT_MODEL).strip()
-    is_openai = model_name == "chat-latest" or model_name.startswith("gpt-")
+    # vendor/model names (e.g. deepseek/deepseek-chat) target an OpenAI-compatible
+    # gateway such as OpenRouter; set OPENAI_BASE_URL and OPENAI_API_KEY for it.
+    is_openai = (
+        model_name == "chat-latest"
+        or model_name.startswith("gpt-")
+        or "/" in model_name
+    )
     if not is_openai and not model_name.startswith("claude-"):
         raise ValueError(
-            "LLM_MODEL must name a supported provider: claude-* (Anthropic) "
-            "or gpt-*/chat-latest (OpenAI). Unset LLM_MODEL to use the default."
+            "LLM_MODEL must name a supported provider: claude-* (Anthropic), "
+            "gpt-*/chat-latest (OpenAI), or vendor/model with OPENAI_BASE_URL. "
+            "Unset LLM_MODEL to use the default."
         )
     key_name = "OPENAI_API_KEY" if is_openai else "ANTHROPIC_API_KEY"
     if not os.environ.get(key_name, "").strip():
diff --git a/apps/agent/tests/test_model.py b/apps/agent/tests/test_model.py
index b9fc29f..c530e67 100644
--- a/apps/agent/tests/test_model.py
+++ b/apps/agent/tests/test_model.py
@@ -84,3 +84,12 @@ def test_missing_provider_key_has_actionable_error(monkeypatch, model, key, valu
         monkeypatch.setenv(key, value)
     with pytest.raises(ValueError, match=key):
         build_model()
+
+
+def test_vendor_model_names_route_to_openai_compatible_gateway(monkeypatch):
+    monkeypatch.setenv("LLM_MODEL", "deepseek/deepseek-chat")
+    monkeypatch.setenv("OPENAI_BASE_URL", "https://openrouter.ai/api/v1")
+    model = build_model()
+    assert isinstance(model, ChatOpenAI)
+    assert model.model_name == "deepseek/deepseek-chat"
+    assert str(model.root_client.base_url).startswith("https://openrouter.ai/api/v1")
diff --git a/docs/deployment.md b/docs/deployment.md
index 4dbb8a9..b069160 100644
--- a/docs/deployment.md
+++ b/docs/deployment.md
@@ -15,7 +15,7 @@ For shared credentials, the blueprint and Python factory default to `LLM_MODEL=c
 
 | Variable                                  | Service       | Purpose                                                                             |
 | ----------------------------------------- | ------------- | ----------------------------------------------------------------------------------- |
-| `LLM_MODEL`                               | Agent         | `claude-*` routes to Anthropic; `gpt-*` routes to OpenAI                            |
+| `LLM_MODEL`                               | Agent         | `claude-*` routes to Anthropic; `gpt-*` to OpenAI; `vendor/model` to `OPENAI_BASE_URL` |
 | `ANTHROPIC_API_KEY`                       | Agent         | Required when selecting Anthropic                                                   |
 | `OPENAI_API_KEY`                          | Agent         | Required when selecting OpenAI                                                      |
 | `LANGSMITH_API_KEY`                       | As configured | Optional tracing configuration                                                      |
diff --git a/docs/getting-started.md b/docs/getting-started.md
index 73fa6fc..41af946 100644
--- a/docs/getting-started.md
+++ b/docs/getting-started.md
@@ -25,7 +25,7 @@ TYPESAFE_API_KEY=your-typesafe-key
 JEV_MODEL=jev-latest
 ```
 
-The model factory accepts `claude-*` names through Anthropic and `chat-latest` or `gpt-*` names through OpenAI. To override the default, set a model available to your account in `LLM_MODEL` and supply its provider key. Shared mode requires the selected answer provider’s key and the Jev key; BYOK requests supply their own pair. Unset `LLM_MODEL` uses the local default; an empty value or unsupported prefix fails with a configuration error.
+The model factory accepts `claude-*` names through Anthropic and `chat-latest` or `gpt-*` names through OpenAI. `vendor/model` names, such as `deepseek/deepseek-chat`, go through an OpenAI-compatible gateway such as OpenRouter: set `OPENAI_BASE_URL=https://openrouter.ai/api/v1` and put the gateway key in `OPENAI_API_KEY`. Browser-supplied keys always use OpenAI. To override the default, set a model available to your account in `LLM_MODEL` and supply its provider key. Shared mode requires the selected answer provider’s key and the Jev key; BYOK requests supply their own pair. Unset `LLM_MODEL` uses the local default; an empty value or unsupported prefix fails with a configuration error.
 
 Check model availability and access with a real request. This repository does not guarantee that every model name works or implement a separate native GPT-6 API. Never put shared provider keys in public frontend environment variables. See [BYOK behavior](../README.md#use-your-own-api-keys).