fix(platform): drop temperature for models that reject it (#1834)

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: Chibionos

packages/uipath-platform/src/uipath/platform/chat/_llm_gateway_service.py

@@ -29,6 +29,7 @@
 from ..common._endpoints_manager import EndpointManager
 from ..common._execution_context import UiPathExecutionContext
 from ..common._models import Endpoint
+from ._model_capabilities import should_skip_temperature
 from .llm_gateway import (
     ChatCompletion,
     SpecificToolChoice,
@@ -175,6 +176,7 @@ def __init__(
         action_id: Optional[str] = None,
     ) -> None:
         super().__init__(config=config, execution_context=execution_context)
+        self._agenthub_config = agenthub_config
         self._llm_headers = _build_llm_headers(
             requesting_product, requesting_feature, agenthub_config, action_id
         )
@@ -329,13 +331,18 @@ class Country(BaseModel):
 
         is_reasoning_model = model.lower().startswith(("o1", "o3", "o4"))
 
+        # Reasoning models (o1, o3, o4) don't support temperature; newer models
+        # reject it too, which only discovery knows about.
+        skip_temperature = is_reasoning_model or await should_skip_temperature(
+            self, model, self._agenthub_config
+        )
+
         request_body: dict[str, Any] = {
             "messages": messages,
             "max_tokens": max_tokens,
         }
 
-        # Reasoning models (o1, o3, o4) don't support temperature
-        if not is_reasoning_model:
+        if not skip_temperature:
             request_body["temperature"] = temperature
 
         # Handle response_format - convert BaseModel to schema if needed
@@ -392,6 +399,7 @@ def __init__(
         action_id: Optional[str] = None,
     ) -> None:
         super().__init__(config=config, execution_context=execution_context)
+        self._agenthub_config = agenthub_config
         self._llm_headers = _build_llm_headers(
             requesting_product, requesting_feature, agenthub_config, action_id
         )
@@ -558,13 +566,18 @@ class Country(BaseModel):
         is_claude_model = "claude" in model_lower
         is_reasoning_model = model_lower.startswith(("o1", "o3", "o4"))
 
+        # Reasoning models (o1, o3, o4) don't support temperature; newer models
+        # reject it too, which only discovery knows about.
+        skip_temperature = is_reasoning_model or await should_skip_temperature(
+            self, model, self._agenthub_config
+        )
+
         request_body: dict[str, Any] = {
             "messages": converted_messages,
             "max_tokens": max_tokens,
         }
 
-        # Reasoning models (o1, o3, o4) don't support temperature and sampling parameters
-        if not is_reasoning_model:
+        if not skip_temperature:
             request_body["temperature"] = temperature
 
         # Only add OpenAI-specific parameters for non-Claude and non-reasoning models

packages/uipath-platform/src/uipath/platform/chat/_model_capabilities.py

@@ -0,0 +1,98 @@
+"""Per-model parameter support, resolved from the LLM Gateway discovery API.
+
+Newer models reject parameters that older ones accept: passing `temperature` to
+a model that dropped it fails the whole request with HTTP 400. The gateway
+publishes these constraints per model under `modelDetails`, so callers can strip
+the parameter instead of guessing from the model name.
+"""
+
+import logging
+from typing import Any
+
+from ..common._base_service import BaseService
+from ..common._endpoints_manager import EndpointManager
+from ..common._models import Endpoint
+from ..constants import HEADER_AGENTHUB_CONFIG
+
+logger = logging.getLogger(__name__)
+
+CacheKey = tuple[str, str | None]
+
+# One fetch per (base_url, agenthub_config); an empty mapping caches "couldn't read
+# discovery". Deliberately unlocked: a racing duplicate GET is cheaper than an
+# asyncio.Lock pinned to whichever event loop touched it first.
+_model_details: dict[CacheKey, dict[str, dict[str, Any]]] = {}
+
+
+async def should_skip_temperature(
+    service: BaseService, model: str, agenthub_config: str | None
+) -> bool:
+    """Whether `model` rejects the `temperature` parameter.
+
+    Args:
+        service: Service used to call discovery; supplies base URL and auth.
+        model: Model name as sent to the gateway.
+        agenthub_config: AgentHub config scoping which models are visible.
+
+    Returns:
+        True only when discovery explicitly reports the model skips temperature.
+        Unknown models, unreachable discovery, and any unexpected failure return
+        False, leaving the caller's request untouched.
+    """
+    try:
+        details = await _details_for(service, model, agenthub_config)
+    except Exception as e:
+        logger.warning(
+            "Could not resolve parameter support for model %s (%s); "
+            "sending model parameters unchanged",
+            model,
+            e,
+        )
+        return False
+
+    return bool(details.get("shouldSkipTemperature", False))
+
+
+async def _details_for(
+    service: BaseService, model: str, agenthub_config: str | None
+) -> dict[str, Any]:
+    key = (service._config.base_url, agenthub_config)
+
+    if key not in _model_details:
+        _model_details[key] = await _fetch_model_details(service, agenthub_config)
+
+    return _model_details[key].get(model, {})
+
+
+async def _fetch_model_details(
+    service: BaseService, agenthub_config: str | None
+) -> dict[str, dict[str, Any]]:
+    headers = {HEADER_AGENTHUB_CONFIG: agenthub_config} if agenthub_config else {}
+    endpoint = Endpoint("/" + EndpointManager.get_discovery_endpoint())
+
+    try:
+        response = await service.request_async("GET", endpoint, headers=headers)
+        models = response.json()
+    except Exception as e:
+        # Falling back to the caller's parameters is the pre-existing behaviour;
+        # failing the LLM call because discovery is down would be worse.
+        logger.warning(
+            "LLM Gateway discovery unavailable (%s); sending model parameters unchanged",
+            e,
+        )
+        return {}
+
+    if not isinstance(models, list):
+        logger.warning("Unexpected LLM Gateway discovery payload; expected a list")
+        return {}
+
+    return {
+        model["modelName"]: model.get("modelDetails") or {}
+        for model in models
+        if isinstance(model, dict) and model.get("modelName")
+    }
+
+
+def _reset_cache() -> None:
+    """Clear the discovery cache. For tests."""
+    _model_details.clear()

packages/uipath-platform/src/uipath/platform/common/_endpoints_manager.py

@@ -21,6 +21,7 @@ class UiPathEndpoints(Enum):
         "agenthub_/llm/raw/vendor/{vendor}/model/{model}/completions"
     )
     AH_CAPABILITIES_ENDPOINT = "agenthub_/llm/api/capabilities"
+    AH_DISCOVERY_ENDPOINT = "agenthub_/llm/api/discovery"
 
     OR_NORMALIZED_COMPLETION_ENDPOINT = "orchestrator_/llm/api/chat/completions"
     OR_PASSTHROUGH_COMPLETION_ENDPOINT = "orchestrator_/llm/openai/deployments/{model}/chat/completions?api-version={api_version}"
@@ -29,6 +30,7 @@ class UiPathEndpoints(Enum):
         "orchestrator_/llm/raw/vendor/{vendor}/model/{model}/completions"
     )
     OR_CAPABILITIES_ENDPOINT = "orchestrator_/llm/api/capabilities"
+    OR_DISCOVERY_ENDPOINT = "orchestrator_/llm/api/discovery"
 
 
 class EndpointManager:
@@ -185,6 +187,14 @@ def get_normalized_endpoint(cls) -> str:
             UiPathEndpoints.OR_NORMALIZED_COMPLETION_ENDPOINT,
         )
 
+    @classmethod
+    def get_discovery_endpoint(cls) -> str:
+        """Get the model discovery endpoint."""
+        return cls._select_endpoint(
+            UiPathEndpoints.AH_DISCOVERY_ENDPOINT,
+            UiPathEndpoints.OR_DISCOVERY_ENDPOINT,
+        )
+
     @classmethod
     def get_embeddings_endpoint(cls) -> str:
         """Get the embeddings endpoint."""