fix: normalize reasoning effort ordering in UI

batch_runner.py

@@ -1158,7 +1158,7 @@ def main(
         providers_order (str): Comma-separated list of OpenRouter providers to try in order (e.g. "anthropic,openai,google")
         provider_sort (str): Sort providers by "price", "throughput", or "latency" (OpenRouter only)
         max_tokens (int): Maximum tokens for model responses (optional, uses model default if not set)
-        reasoning_effort (str): OpenRouter reasoning effort level: "xhigh", "high", "medium", "low", "minimal", "none" (default: "medium")
+        reasoning_effort (str): OpenRouter reasoning effort level: "none", "minimal", "low", "medium", "high", "xhigh" (default: "medium")
         reasoning_disabled (bool): Completely disable reasoning/thinking tokens (default: False)
         prefill_messages_file (str): Path to JSON file containing prefill messages (list of {role, content} dicts)
         max_samples (int): Only process the first N samples from the dataset (optional, processes all if not set)
@@ -1227,7 +1227,7 @@ def main(
         print("🧠 Reasoning: DISABLED (effort=none)")
     elif reasoning_effort:
         # Use specified effort level
-        valid_efforts = ["xhigh", "high", "medium", "low", "minimal", "none"]
+        valid_efforts = ["none", "minimal", "low", "medium", "high", "xhigh"]
         if reasoning_effort not in valid_efforts:
             print(f"❌ Error: --reasoning_effort must be one of: {', '.join(valid_efforts)}")
             return

cli.py

@@ -5259,7 +5259,7 @@ class HermesCLI:
 
         Usage:
             /reasoning              Show current effort level and display state
-            /reasoning <level>      Set reasoning effort (none, low, medium, high, xhigh)
+            /reasoning <level>      Set reasoning effort (none, minimal, low, medium, high, xhigh)
             /reasoning show|on      Show model thinking/reasoning in output
             /reasoning hide|off     Hide model thinking/reasoning from output
         """
@@ -5277,7 +5277,7 @@ class HermesCLI:
             display_state = "on ✓" if self.show_reasoning else "off"
             _cprint(f"  {_GOLD}Reasoning effort:  {level}{_RST}")
             _cprint(f"  {_GOLD}Reasoning display: {display_state}{_RST}")
-            _cprint(f"  {_DIM}Usage: /reasoning <none|low|medium|high|xhigh|show|hide>{_RST}")
+            _cprint(f"  {_DIM}Usage: /reasoning <none|minimal|low|medium|high|xhigh|show|hide>{_RST}")
             return
 
         arg = parts[1].strip().lower()
@@ -5303,7 +5303,7 @@ class HermesCLI:
         parsed = _parse_reasoning_config(arg)
         if parsed is None:
             _cprint(f"  {_DIM}(._.) Unknown argument: {arg}{_RST}")
-            _cprint(f"  {_DIM}Valid levels: none, low, minimal, medium, high, xhigh{_RST}")
+            _cprint(f"  {_DIM}Valid levels: none, minimal, low, medium, high, xhigh{_RST}")
             _cprint(f"  {_DIM}Display:      show, hide{_RST}")
             return
 

gateway/platforms/discord.py

@@ -1595,7 +1595,7 @@ class DiscordAdapter(BasePlatformAdapter):
             await self._run_simple_slash(interaction, f"/model {name}".strip())
 
         @tree.command(name="reasoning", description="Show or change reasoning effort")
-        @discord.app_commands.describe(effort="Reasoning effort: xhigh, high, medium, low, minimal, or none.")
+        @discord.app_commands.describe(effort="Reasoning effort: none, minimal, low, medium, high, or xhigh.")
         async def slash_reasoning(interaction: discord.Interaction, effort: str = ""):
             await self._run_simple_slash(interaction, f"/reasoning {effort}".strip())
 

gateway/run.py

@@ -4840,7 +4840,7 @@ class GatewayRunner:
 
         Usage:
             /reasoning              Show current effort level and display state
-            /reasoning <level>      Set reasoning effort (none, low, medium, high, xhigh)
+            /reasoning <level>      Set reasoning effort (none, minimal, low, medium, high, xhigh)
             /reasoning show|on      Show model reasoning in responses
             /reasoning hide|off     Hide model reasoning from responses
         """
@@ -4885,7 +4885,7 @@ class GatewayRunner:
                 "🧠 **Reasoning Settings**\n\n"
                 f"**Effort:** `{level}`\n"
                 f"**Display:** {display_state}\n\n"
-                "_Usage:_ `/reasoning <none|low|medium|high|xhigh|show|hide>`"
+                "_Usage:_ `/reasoning <none|minimal|low|medium|high|xhigh|show|hide>`"
             )
 
         # Display toggle
@@ -4908,7 +4908,7 @@ class GatewayRunner:
         else:
             return (
                 f"⚠️ Unknown argument: `{effort}`\n\n"
-                "**Valid levels:** none, low, minimal, medium, high, xhigh\n"
+                "**Valid levels:** none, minimal, low, medium, high, xhigh\n"
                 "**Display:** show, hide"
             )
 

hermes_cli/commands.py

@@ -99,7 +99,7 @@ COMMAND_REGISTRY: list[CommandDef] = [
                "Configuration"),
     CommandDef("reasoning", "Manage reasoning effort and display", "Configuration",
                args_hint="[level|show|hide]",
-               subcommands=("none", "low", "minimal", "medium", "high", "xhigh", "show", "hide", "on", "off")),
+               subcommands=("none", "minimal", "low", "medium", "high", "xhigh", "show", "hide", "on", "off")),
     CommandDef("skin", "Show or change the display skin/theme", "Configuration",
                cli_only=True, args_hint="[name]"),
     CommandDef("voice", "Toggle voice mode", "Configuration",

hermes_cli/main.py

@@ -1811,7 +1811,10 @@ def _set_reasoning_effort(config, effort: str) -> None:
 
 def _prompt_reasoning_effort_selection(efforts, current_effort=""):
     """Prompt for a reasoning effort. Returns effort, 'none', or None to keep current."""
-    ordered = list(dict.fromkeys(str(effort).strip().lower() for effort in efforts if str(effort).strip()))
+    deduped = list(dict.fromkeys(str(effort).strip().lower() for effort in efforts if str(effort).strip()))
+    canonical_order = ("minimal", "low", "medium", "high", "xhigh")
+    ordered = [effort for effort in canonical_order if effort in deduped]
+    ordered.extend(effort for effort in deduped if effort not in canonical_order)
     if not ordered:
         return None
 

hermes_constants.py

@@ -78,7 +78,7 @@ VALID_REASONING_EFFORTS = ("xhigh", "high", "medium", "low", "minimal")
 def parse_reasoning_effort(effort: str) -> dict | None:
     """Parse a reasoning effort level into a config dict.
 
-    Valid levels: "xhigh", "high", "medium", "low", "minimal", "none".
+    Valid levels: "none", "minimal", "low", "medium", "high", "xhigh".
     Returns None when the input is empty or unrecognized (caller uses default).
     Returns {"enabled": False} for "none".
     Returns {"enabled": True, "effort": <level>} for valid effort levels.

tests/hermes_cli/test_commands.py

@@ -68,6 +68,17 @@ class TestCommandRegistry:
         for cmd in COMMAND_REGISTRY:
             assert cmd.category in valid_categories, f"{cmd.name} has invalid category '{cmd.category}'"
 
+    def test_reasoning_subcommands_are_in_logical_order(self):
+        reasoning = next(cmd for cmd in COMMAND_REGISTRY if cmd.name == "reasoning")
+        assert reasoning.subcommands[:6] == (
+            "none",
+            "minimal",
+            "low",
+            "medium",
+            "high",
+            "xhigh",
+        )
+
     def test_cli_only_and_gateway_only_are_mutually_exclusive(self):
         for cmd in COMMAND_REGISTRY:
             assert not (cmd.cli_only and cmd.gateway_only), \

tests/hermes_cli/test_reasoning_effort_menu.py

@@ -0,0 +1,34 @@
+import sys
+import types
+
+
+from hermes_cli.main import _prompt_reasoning_effort_selection
+
+
+class _FakeTerminalMenu:
+    last_choices = None
+
+    def __init__(self, choices, **kwargs):
+        _FakeTerminalMenu.last_choices = choices
+        self._cursor_index = kwargs.get("cursor_index")
+
+    def show(self):
+        return self._cursor_index
+
+
+def test_reasoning_menu_orders_minimal_before_low(monkeypatch):
+    fake_module = types.SimpleNamespace(TerminalMenu=_FakeTerminalMenu)
+    monkeypatch.setitem(sys.modules, "simple_term_menu", fake_module)
+
+    selected = _prompt_reasoning_effort_selection(
+        ["low", "minimal", "medium", "high"],
+        current_effort="medium",
+    )
+
+    assert selected == "medium"
+    assert _FakeTerminalMenu.last_choices[:4] == [
+        "  minimal",
+        "  low",
+        "  medium  ← currently in use",
+        "  high",
+    ]