diff --git a/.env.example b/.env.example
index 454713d..cba2ef7 100644
--- a/.env.example
+++ b/.env.example
@@ -1,4 +1,5 @@
 OLLAMA_HOST=http://localhost:11434
+OLLAMA_API_KEY=
 OLLAMA_DEFAULT_MODEL=gemma4:e4b-it-q_8
 
 OPENAI_API_KEY=
diff --git a/README.md b/README.md
index a32a207..ec36d0f 100644
--- a/README.md
+++ b/README.md
@@ -142,6 +142,7 @@
 ```dotenv
 # LLM
 OLLAMA_HOST=http://localhost:11434
+OLLAMA_API_KEY=
 OLLAMA_DEFAULT_MODEL=gemma4:e4b-it-q8_0
 OLLAMA_NUM_CTX=65536
 OLLAMA_THINK=true
@@ -153,3 +154,12 @@
 # Персона
 NAVI_PERSONA_FILE=persona.txt
 ```
+
+Для прямого доступа к Ollama Cloud:
+
+```dotenv
+OLLAMA_HOST=https://ollama.com
+OLLAMA_API_KEY=your_api_key
+```
+
+Также укажи cloud-модель в `navi/profiles/*/config.json`, потому что профили могут переопределять `OLLAMA_DEFAULT_MODEL`.
diff --git a/docs/config.md b/docs/config.md
index 86bdc31..9dabc4c 100644
--- a/docs/config.md
+++ b/docs/config.md
@@ -7,12 +7,18 @@
 | Variable | Type | Default | Description |
 |---|---|---|---|
 | `OLLAMA_HOST` | str | `http://localhost:11434` | Ollama server URL |
+| `OLLAMA_API_KEY` | str | `""` | Ollama Cloud API key for direct `https://ollama.com` access |
 | `OLLAMA_DEFAULT_MODEL` | str | `gemma4:e2b-it-q8_0` | Default model (can be overridden per profile) |
 | `OLLAMA_NUM_CTX` | int | `65536` | Context window size in tokens |
 | `OLLAMA_THINK` | bool | `true` | Enable extended reasoning (thinking) |
 | `OPENAI_API_KEY` | str | `""` | OpenAI API key (if using OpenAI backend) |
 | `ANTHROPIC_API_KEY` | str | `""` | Anthropic API key (if using Anthropic backend) |
 
+For direct Ollama Cloud access, set `OLLAMA_HOST=https://ollama.com` and
+`OLLAMA_API_KEY=<key>`. Built-in profiles may override `OLLAMA_DEFAULT_MODEL`,
+so update the `model` field in `navi/profiles/*/config.json` to a cloud model
+when switching away from local Ollama models.
+
 ## Security / Sandboxing
 
 | Variable | Type | Default | Description |
@@ -73,6 +79,7 @@
 
 ```dotenv
 OLLAMA_HOST=http://localhost:11434
+OLLAMA_API_KEY=
 OLLAMA_DEFAULT_MODEL=gemma4:e2b-it-q8_0
 OLLAMA_NUM_CTX=65536
 OLLAMA_THINK=true
diff --git a/navi/config.py b/navi/config.py
index 05a3a63..65f9f4a 100644
--- a/navi/config.py
+++ b/navi/config.py
@@ -8,6 +8,7 @@
     model_config = SettingsConfigDict(env_file=".env", env_file_encoding="utf-8", extra="ignore")
 
     ollama_host: str = "http://localhost:11434"
+    ollama_api_key: str = ""
     ollama_default_model: str = "gemma4:e4b-it-q8_0"
     ollama_num_ctx: int = 65536
     ollama_think: bool = True
diff --git a/navi/core/registry.py b/navi/core/registry.py
index a768eee..56d377f 100644
--- a/navi/core/registry.py
+++ b/navi/core/registry.py
@@ -112,6 +112,7 @@
     ollama_backend = OllamaBackend(
         model=settings.ollama_default_model,
         host=settings.ollama_host,
+        api_key=settings.ollama_api_key,
     )
     ai_helper = AIHelper(
         backend=ollama_backend,
diff --git a/navi/llm/ollama.py b/navi/llm/ollama.py
index cee4e8d..6f6bfef 100644
--- a/navi/llm/ollama.py
+++ b/navi/llm/ollama.py
@@ -46,9 +46,10 @@
 
 
 class OllamaBackend(LLMBackend):
-    def __init__(self, model: str, host: str = "http://localhost:11434"):
+    def __init__(self, model: str, host: str = "http://localhost:11434", api_key: str = ""):
         self.model = model
-        self._client = ollama_client.AsyncClient(host=host)
+        headers = {"Authorization": f"Bearer {api_key}"} if api_key else None
+        self._client = ollama_client.AsyncClient(host=host, headers=headers)
 
     async def complete(
         self,
diff --git a/navi/profiles/developer/config.json b/navi/profiles/developer/config.json
index cb115dc..a320cf5 100644
--- a/navi/profiles/developer/config.json
+++ b/navi/profiles/developer/config.json
@@ -9,7 +9,7 @@
     "key_tools": "filesystem, code_exec, terminal, web_search, web_view, spawn_agent"
   },
   "llm_backend": "ollama",
-  "model": "gemma4:26b-a4b-it-q4_K_M",
+  "model": "gemma4:31b",
   "temperature": 0.2,
   "max_iterations": 35,
   "planning_enabled": true,
diff --git a/navi/profiles/secretary/config.json b/navi/profiles/secretary/config.json
index 5ee8e98..fef7db3 100644
--- a/navi/profiles/secretary/config.json
+++ b/navi/profiles/secretary/config.json
@@ -9,7 +9,7 @@
     "key_tools": "web_search, web_view, filesystem, code_exec, gmail, todo, scratchpad, spawn_agent, memory"
   },
   "llm_backend": "ollama",
-  "model": "gemma4:26b-a4b-it-q4_K_M",
+  "model": "gemma4:31b",
   "temperature": 0.5,
   "max_iterations": 25,
   "planning_enabled": true,
diff --git a/navi/profiles/server_admin/config.json b/navi/profiles/server_admin/config.json
index f37228c..e6ff0ce 100644
--- a/navi/profiles/server_admin/config.json
+++ b/navi/profiles/server_admin/config.json
@@ -9,7 +9,7 @@
     "key_tools": "ssh_exec, terminal, filesystem, code_exec, web_search, spawn_agent, memory"
   },
   "llm_backend": "ollama",
-  "model": "gemma4:26b-a4b-it-q4_K_M",
+  "model": "gemma4:31b",
   "temperature": 0.2,
   "max_iterations": 20,
   "planning_enabled": true,
diff --git a/navi/profiles/tool_developer/config.json b/navi/profiles/tool_developer/config.json
index 75e0375..b89c274 100644
--- a/navi/profiles/tool_developer/config.json
+++ b/navi/profiles/tool_developer/config.json
@@ -9,7 +9,7 @@
     "key_tools": "write_tool, reload_tools, delete_tool, test_tool, filesystem, terminal, code_exec, memory"
   },
   "llm_backend": "ollama",
-  "model": "gemma4:26b-a4b-it-q4_K_M",
+  "model": "gemma4:31b",
   "temperature": 0.2,
   "max_iterations": 35,
   "planning_enabled": true,