feat(x_search): default model grok-4.20-reasoning -> grok-4.5 (#67719)

grok-4.5 is xAI's newest release (their versioning is non-monotonic:
4.5 > 4.20) and is the model xAI's own docs use for the server-side
x_search tool. Users who explicitly pinned x_search.model keep their
choice; everyone else picks up the new default via the config
deep-merge — no _config_version bump needed.

- tools/x_search_tool.py: DEFAULT_X_SEARCH_MODEL
- hermes_cli/config.py: DEFAULT_CONFIG x_search.model + comment
- agent/reasoning_timeouts.py: 300s stale-timeout floor entry for
  grok-4.5 (grok-4.20-reasoning entry kept for pinned users)
- docs: x-search.md en + zh-Hans (config sample + troubleshooting)
- tests: default-model assertion + timeout-floor positive case

agent/reasoning_timeouts.py

@@ -111,6 +111,7 @@ _REASONING_STALE_TIMEOUT_FLOORS: tuple[tuple[str, int], ...] = (
     # non-reasoning pairs.
     ("grok-4-fast-reasoning", 300),
     ("grok-4.20-reasoning", 300),
+    ("grok-4.5", 300),
     ("grok-4-fast-non-reasoning", 180),
 )
 

hermes_cli/config.py

@@ -3299,10 +3299,10 @@ DEFAULT_CONFIG = {
     # OAuth or XAI_API_KEY) AND the x_search toolset is enabled in
     # `hermes tools`. These settings tune the backing Responses API call.
     "x_search": {
-        # xAI model used for the Responses call. grok-4.20-reasoning is
-        # the recommended default; any Grok model with x_search tool
+        # xAI model used for the Responses call. grok-4.5 is the
+        # recommended default; any Grok model with x_search tool
         # access works.
-        "model": "grok-4.20-reasoning",
+        "model": "grok-4.5",
         # Request timeout in seconds (minimum 30). x_search can take
         # 60-120s for complex queries — the default is generous.
         "timeout_seconds": 180,

tests/agent/test_reasoning_stale_timeout_floor.py

@@ -77,6 +77,7 @@ import pytest
     # xAI Grok reasoning variants — explicit, not bare `grok`.
     ("x-ai/grok-4-fast-reasoning", 300.0),
     ("x-ai/grok-4.20-reasoning", 300.0),
+    ("x-ai/grok-4.5", 300.0),
     ("x-ai/grok-4-fast-non-reasoning", 180.0),
 ])
 def test_reasoning_stale_timeout_floor_positive_cases(model, expected):

tests/tools/test_x_search_tool.py

@@ -70,7 +70,7 @@ def test_x_search_posts_responses_request(monkeypatch):
     tool_def = captured["json"]["tools"][0]
     assert captured["url"] == "https://api.x.ai/v1/responses"
     assert captured["headers"]["User-Agent"] == f"Hermes-Agent/{__version__}"
-    assert captured["json"]["model"] == "grok-4.20-reasoning"
+    assert captured["json"]["model"] == "grok-4.5"
     assert captured["json"]["store"] is False
     assert tool_def["type"] == "x_search"
     assert tool_def["allowed_x_handles"] == ["xai", "grok"]

tools/x_search_tool.py

@@ -56,7 +56,7 @@ from tools.xai_http import hermes_xai_user_agent, resolve_xai_http_credentials
 logger = logging.getLogger(__name__)
 
 DEFAULT_XAI_BASE_URL = "https://api.x.ai/v1"
-DEFAULT_X_SEARCH_MODEL = "grok-4.20-reasoning"
+DEFAULT_X_SEARCH_MODEL = "grok-4.5"
 DEFAULT_X_SEARCH_TIMEOUT_SECONDS = 180
 DEFAULT_X_SEARCH_RETRIES = 2
 MAX_HANDLES = 10

website/docs/user-guide/features/x-search.md

@@ -50,9 +50,9 @@ Either choice satisfies the gating. You can pick whichever credentials you alrea
 # ~/.hermes/config.yaml
 x_search:
   # xAI model used for the Responses call.
-  # grok-4.20-reasoning is the recommended default; any Grok model
+  # grok-4.5 is the recommended default; any Grok model
   # with x_search tool access works.
-  model: grok-4.20-reasoning
+  model: grok-4.5
 
   # Request timeout in seconds. x_search can take 60–120s for
   # complex queries — the default is generous. Minimum: 30.
@@ -118,7 +118,7 @@ The tool surfaces this when both auth paths fail. Either set `XAI_API_KEY` in `~
 
 ### "`x_search` is not enabled for this model"
 
-The configured `x_search.model` doesn't have access to the server-side `x_search` tool. Switch to `grok-4.20-reasoning` (the default) or another Grok model that supports it. Check the [xAI documentation](https://docs.x.ai/) for the current list.
+The configured `x_search.model` doesn't have access to the server-side `x_search` tool. Switch to `grok-4.5` (the default) or another Grok model that supports it. Check the [xAI documentation](https://docs.x.ai/) for the current list.
 
 ### Tool doesn't appear in the schema
 

website/i18n/zh-Hans/docusaurus-plugin-content-docs/current/user-guide/features/x-search.md

@@ -46,9 +46,9 @@ hermes tools
 # ~/.hermes/config.yaml
 x_search:
   # 用于 Responses 调用的 xAI 模型。
-  # grok-4.20-reasoning 是推荐的默认值；任何支持
+  # grok-4.5 是推荐的默认值；任何支持
   # x_search 工具访问权限的 Grok 模型均可使用。
-  model: grok-4.20-reasoning
+  model: grok-4.5
 
   # 请求超时时间（秒）。复杂查询的 x_search 可能需要 60–120 秒，
   # 默认值较为宽松。最小值：30。
@@ -114,7 +114,7 @@ agent 将：
 
 ### "`x_search` is not enabled for this model"
 
-配置的 `x_search.model` 没有访问服务端 `x_search` 工具的权限。请切换至 `grok-4.20-reasoning`（默认值）或其他支持该工具的 Grok 模型。当前支持列表请查阅 [xAI 文档](https://docs.x.ai/)。
+配置的 `x_search.model` 没有访问服务端 `x_search` 工具的权限。请切换至 `grok-4.5`（默认值）或其他支持该工具的 Grok 模型。当前支持列表请查阅 [xAI 文档](https://docs.x.ai/)。
 
 ### 工具未出现在 schema 中