lightspeed-stack 0.7.1__tar.gz → 0.7.2__tar.gz

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: lightspeed-stack
-Version: 0.7.1
+Version: 0.7.2
 Summary: LLM tooling stack
 Keywords: LLM,RAG
 Maintainer-Email: Pavel Tišnovský <tisnik@centrum.cz>
@@ -501,11 +501,11 @@ __Note__: Support for individual models is dependent on the specific inference p
 
 | Provider       | Model                                                                        | Tool Calling  | provider_type    | Example                                                                    |
 |----------------|------------------------------------------------------------------------------|---------------|------------------|----------------------------------------------------------------------------|
-| OpenAI         | gpt-5, gpt-4o, gpt-4-turbo, gpt-4.1, o1, o3, o4                               | Yes           | remote::openai   | [1](examples/openai-faiss-run.yaml) [2](examples/openai-pgvector-run.yaml) |
+| OpenAI         | gpt-5, gpt-5.6-luna, gpt-4-turbo, gpt-4.1, o1, o3, o4                               | Yes           | remote::openai   | [1](examples/openai-faiss-run.yaml) [2](examples/openai-pgvector-run.yaml) |
 | OpenAI         | gpt-3.5-turbo, gpt-4                                                         | No            | remote::openai   |                                                                            |
 | RHOAI (vLLM)   | meta-llama/Llama-3.2-1B-Instruct                                             | Yes           | remote::vllm     | [1](tests/e2e-prow/rhoai/configs/run.yaml)                                 |
 | RHEL AI (RHAIIS/vLLM) | meta-llama/Llama-3.1-8B-Instruct                                      | Yes           | remote::vllm     | [1](tests/e2e/configs/run-rhelai.yaml) [2](tests/e2e/configs/run-rhaiis.yaml) |
-| Azure          | gpt-5, gpt-5-mini, gpt-5-nano, gpt-4o-mini, o3-mini, o4-mini, o1             | Yes           | remote::azure    | [1](examples/azure-run.yaml)                                               |
+| Azure          | gpt-5, gpt-5-mini, gpt-5-nano, gpt-5.6-luna, o3-mini, o4-mini, o1             | Yes           | remote::azure    | [1](examples/azure-run.yaml)                                               |
 | Azure          | gpt-5-chat, gpt-4.1, gpt-4.1-mini, gpt-4.1-nano,  o1-mini                    | No or limited | remote::azure    |                                                                            |
 | VertexAI       | google/gemini-2.0-flash, google/gemini-2.5-flash, google/gemini-2.5-pro [^1] | Yes           | remote::vertexai | [1](examples/vertexai-run.yaml)                                            |
 | WatsonX        | meta-llama/llama-3-3-70b-instruct                                            | Yes           | remote::watsonx  | [1](examples/watsonx-run.yaml)                                             |
@@ -1514,12 +1514,12 @@ will be returned.
       "model_type": "embedding"
     },
     {
-      "identifier": "openai/gpt-4o-mini",
+      "identifier": "openai/gpt-5.6-luna",
       "metadata": {},
       "api_model_type": "llm",
       "provider_id": "openai",
       "type": "model",
-      "provider_resource_id": "gpt-4o-mini",
+      "provider_resource_id": "gpt-5.6-luna",
       "model_type": "llm"
     },
     {

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/README.md

@@ -241,11 +241,11 @@ __Note__: Support for individual models is dependent on the specific inference p
 
 | Provider       | Model                                                                        | Tool Calling  | provider_type    | Example                                                                    |
 |----------------|------------------------------------------------------------------------------|---------------|------------------|----------------------------------------------------------------------------|
-| OpenAI         | gpt-5, gpt-4o, gpt-4-turbo, gpt-4.1, o1, o3, o4                               | Yes           | remote::openai   | [1](examples/openai-faiss-run.yaml) [2](examples/openai-pgvector-run.yaml) |
+| OpenAI         | gpt-5, gpt-5.6-luna, gpt-4-turbo, gpt-4.1, o1, o3, o4                               | Yes           | remote::openai   | [1](examples/openai-faiss-run.yaml) [2](examples/openai-pgvector-run.yaml) |
 | OpenAI         | gpt-3.5-turbo, gpt-4                                                         | No            | remote::openai   |                                                                            |
 | RHOAI (vLLM)   | meta-llama/Llama-3.2-1B-Instruct                                             | Yes           | remote::vllm     | [1](tests/e2e-prow/rhoai/configs/run.yaml)                                 |
 | RHEL AI (RHAIIS/vLLM) | meta-llama/Llama-3.1-8B-Instruct                                      | Yes           | remote::vllm     | [1](tests/e2e/configs/run-rhelai.yaml) [2](tests/e2e/configs/run-rhaiis.yaml) |
-| Azure          | gpt-5, gpt-5-mini, gpt-5-nano, gpt-4o-mini, o3-mini, o4-mini, o1             | Yes           | remote::azure    | [1](examples/azure-run.yaml)                                               |
+| Azure          | gpt-5, gpt-5-mini, gpt-5-nano, gpt-5.6-luna, o3-mini, o4-mini, o1             | Yes           | remote::azure    | [1](examples/azure-run.yaml)                                               |
 | Azure          | gpt-5-chat, gpt-4.1, gpt-4.1-mini, gpt-4.1-nano,  o1-mini                    | No or limited | remote::azure    |                                                                            |
 | VertexAI       | google/gemini-2.0-flash, google/gemini-2.5-flash, google/gemini-2.5-pro [^1] | Yes           | remote::vertexai | [1](examples/vertexai-run.yaml)                                            |
 | WatsonX        | meta-llama/llama-3-3-70b-instruct                                            | Yes           | remote::watsonx  | [1](examples/watsonx-run.yaml)                                             |
@@ -1254,12 +1254,12 @@ will be returned.
       "model_type": "embedding"
     },
     {
-      "identifier": "openai/gpt-4o-mini",
+      "identifier": "openai/gpt-5.6-luna",
       "metadata": {},
       "api_model_type": "llm",
       "provider_id": "openai",
       "type": "model",
-      "provider_resource_id": "gpt-4o-mini",
+      "provider_resource_id": "gpt-5.6-luna",
       "model_type": "llm"
     },
     {

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/pyproject.toml

@@ -63,7 +63,7 @@ dependencies = [
     "opentelemetry-exporter-otlp>=1.34.1",
     "opentelemetry-instrumentation-fastapi>=0.49b0",
 ]
-version = "0.7.1"
+version = "0.7.2"
 
 [project.license]
 file = "LICENSE"

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/app/endpoints/responses.py

@@ -33,7 +33,11 @@ from authorization.azure_token_manager import AzureEntraIDManager
 from authorization.middleware import authorize
 from client.ogx import AsyncOgxClientHolder
 from configuration import configuration
-from constants import ENDPOINT_PATH_RESPONSES, SUBSTITUTED_INSTRUCTIONS_PLACEHOLDER
+from constants import (
+    ENDPOINT_PATH_RESPONSES,
+    RESPONSES_REQUEST_MAX_SIZE,
+    SUBSTITUTED_INSTRUCTIONS_PLACEHOLDER,
+)
 from log import get_logger
 from metrics import recording
 from models.api.requests import ResponsesRequest
@@ -100,7 +104,7 @@ from utils.responses import (
     build_turn_summary,
     check_model_configured,
     deduplicate_referenced_documents,
-    extract_attachments_text,
+    extract_shield_moderation_text,
     extract_text_from_response_items,
     extract_token_usage,
     extract_vector_store_ids_from_tools,
@@ -631,13 +635,18 @@ async def handle_responses_with_tracing(  # pylint: disable=too-many-locals
     ):
         client = await AsyncOgxClientHolder().update_azure_token()
 
-    attachments_text = extract_attachments_text(original_request.input)
     add_span_event(root_span, SpanEvents.VALIDATION_COMPLETED)
 
     endpoint_path = ENDPOINT_PATH_RESPONSES
 
+    shield_after_last_refusal = (
+        request.headers.get("X-LCS-Shield-After-Last-Refusal", "").lower() == "true"
+    )
     moderation_result = await run_shield_moderation_v2(
-        input_text + "\n\n" + attachments_text,
+        extract_shield_moderation_text(
+            original_request.input,
+            shield_after_last_refusal,
+        ),
         configuration.configuration.shields,
         responses_request.shield_ids,
     )
@@ -660,6 +669,7 @@ async def handle_responses_with_tracing(  # pylint: disable=too-many-locals
         original_request.reasoning,
         updated_request.tools,
         updated_request.model,
+        configuration.skills,
     )
 
     # Extract vector store IDs for Inline RAG context from the original request
@@ -1476,6 +1486,13 @@ async def handle_non_streaming_response(
             tools,
             configuration.rag_id_mapping,
         )
+
+    # retrieve max size of response request
+    responses_request_max_size = RESPONSES_REQUEST_MAX_SIZE
+    customization = configuration.configuration.customization
+    if customization is not None:
+        responses_request_max_size = customization.responses_request_max_size
+
     return ResponsesResponse.model_validate(
         {
             **response_dict,
@@ -1483,5 +1500,6 @@ async def handle_non_streaming_response(
             "conversation": normalize_conversation_id(api_params.conversation),
             "completed_at": int(completed_at.timestamp()),
             "output_text": output_text,
-        }
+        },
+        context={"responses_request_max_size": responses_request_max_size},
     )

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/data/default_run.yaml

@@ -39,7 +39,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY:=}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - provider_id: sentence-transformers
     provider_type: inline::sentence-transformers
   files:

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/models/api/requests/responses_openai.py

@@ -15,7 +15,7 @@ from ogx_api.openai_responses import (
 from ogx_api.openai_responses import (
     OpenAIResponseText as Text,
 )
-from pydantic import BaseModel, field_validator, model_validator
+from pydantic import BaseModel, ValidationInfo, field_validator, model_validator
 
 from constants import RESPONSES_REQUEST_MAX_SIZE
 from models.common.query import SolrVectorSearchRequest
@@ -105,7 +105,7 @@ class ResponsesRequest(BaseModel):
 
     @model_validator(mode="before")
     @classmethod
-    def validate_body_size(cls, values: Any) -> Any:
+    def validate_body_size(cls, values: Any, info: ValidationInfo) -> Any:
         """Validate that the request body does not exceed the maximum allowed size.
 
         Serializes the raw request payload to JSON and checks the total character
@@ -130,10 +130,15 @@ class ResponsesRequest(BaseModel):
             # model instances).  The size guard only applies to wire-format HTTP
             # requests which FastAPI always parses into JSON-compatible dicts.
             return values
-        if len(serialized) > RESPONSES_REQUEST_MAX_SIZE:
+
+        max_size = (info.context or {}).get(
+            "responses_request_max_size", RESPONSES_REQUEST_MAX_SIZE
+        )
+
+        if len(serialized) > max_size:
             raise ValueError(
                 f"Request body size ({len(serialized)} characters) exceeds maximum "
-                f"allowed size of {RESPONSES_REQUEST_MAX_SIZE} characters"
+                f"allowed size of {max_size} characters"
             )
         return values
 

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/models/common/moderation.py

@@ -23,10 +23,11 @@ class ShieldModerationBlocked(BaseModel):
 
     @property
     def refusal_response(self) -> ResponseMessage:
-        """Build a ResponseMessage carrying the shield's refusal text."""
+        """Build a refusal message; ``id`` is the moderation id (``modr-…``)."""
         return ResponseMessage(
             role="assistant",
             content=self.message,
+            id=self.moderation_id,
         )
 
 

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/models/config.py

@@ -1716,6 +1716,12 @@ class Customization(ConfigurationBase):
     agent_card_path: Optional[FilePath] = None
     agent_card_config: Optional[dict[str, Any]] = None
     custom_profile: Optional[CustomProfile] = Field(default=None, init=False)
+    responses_request_max_size: int = Field(
+        default=constants.RESPONSES_REQUEST_MAX_SIZE,
+        title="Maximum character length for the serialized responses request body",
+        description="Maximum character length for the serialized /v1/response responses "
+        "request body",
+    )
 
     @model_validator(mode="after")
     def check_customization_model(self) -> Self:

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/src/utils/responses.py

@@ -113,7 +113,7 @@ from models.common.turn_summary import (
     ToolResultSummary,
     TurnSummary,
 )
-from models.config import RagStore
+from models.config import RagStore, SkillsConfiguration
 from models.database.conversations import UserConversation
 from utils.mcp.mcp_headers import (
     McpHeaders,
@@ -135,7 +135,7 @@ from utils.query import (
     normalize_vertex_ai_model_id,
     prepare_input,
 )
-from utils.suid import to_ogx_conversation_id
+from utils.suid import is_moderation_id, to_ogx_conversation_id
 from utils.token_counter import TokenCounter
 
 logger = get_logger(__name__)
@@ -439,7 +439,12 @@ async def prepare_responses_params(  # pylint: disable=too-many-arguments,too-ma
     normalized_model = normalize_vertex_ai_model_id(model)
 
     # Known LLS bug: https://redhat.atlassian.net/browse/LCORE-1583
-    reasoning = apply_reasoning_for_resolved_tools(None, tools, normalized_model)
+    reasoning = apply_reasoning_for_resolved_tools(
+        None,
+        tools,
+        normalized_model,
+        configuration.skills,
+    )
 
     return ResponsesApiParams(
         input=input_text,
@@ -1737,6 +1742,64 @@ def extract_attachments_text(response_input: ResponseInput) -> str:
     return "\n\n".join(file_data_parts)
 
 
+def get_shield_moderation_input(
+    response_input: ResponseInput,
+    after_last_refusal: bool = False,
+) -> ResponseInput:
+    """Return input for shield moderation, optionally after the last refusal.
+
+    Args:
+        response_input: Responses API ``input`` (string or item list).
+        after_last_refusal: When True, window after the last ``modr-…`` item.
+
+    Returns:
+        The full input, or the suffix list after the last moderated refusal.
+
+    """
+    if not after_last_refusal or isinstance(response_input, str):
+        return response_input
+
+    item_count = len(response_input)
+    for offset, item in enumerate(reversed(response_input)):
+        if (
+            isinstance(item, ResponseMessage)
+            and item.role == "assistant"
+            and item.id is not None
+            and is_moderation_id(item.id)
+        ):
+            return list(response_input[item_count - offset :])
+
+    return response_input
+
+
+def extract_shield_moderation_text(
+    response_input: ResponseInput,
+    after_last_refusal: bool = False,
+) -> str:
+    """Extract text and attachments for shield moderation.
+
+    Optionally windows to items after the last ``modr-…`` refusal first.
+
+    Args:
+        response_input: Responses API ``input`` (string or item list).
+        after_last_refusal: When True, window after the last ``modr-…`` item.
+
+    Returns:
+        Combined message text and attachment ``file_data`` for shields.
+
+    """
+    shield_input = get_shield_moderation_input(
+        response_input,
+        after_last_refusal=after_last_refusal,
+    )
+    text = (
+        shield_input
+        if isinstance(shield_input, str)
+        else extract_text_from_response_items(shield_input)
+    )
+    return text + "\n\n" + extract_attachments_text(shield_input)
+
+
 def _merge_tools(
     client_tools: list[InputTool],
     server_tools: list[InputTool],
@@ -1864,6 +1927,7 @@ def apply_reasoning_for_resolved_tools(
     reasoning: Optional[OpenAIResponseReasoning],
     tools: Optional[list[InputTool]],
     model_id: str,
+    skills_configuration: Optional[SkillsConfiguration],
 ) -> Optional[OpenAIResponseReasoning]:
     """Set reasoning effort to ``none`` when resolved tools are present.
 
@@ -1875,13 +1939,15 @@ def apply_reasoning_for_resolved_tools(
         reasoning: Optional reasoning configuration to modify.
         tools: Optional list of tools that have been resolved.
         model_id: Resolved model identifier in ``provider/model`` form.
-
+        skills_configuration: Optional skills configuration.
     Returns:
         Modified reasoning configuration with effort set to "none" if tools
         are present on a default-on reasoning model, or the original reasoning
         otherwise.
     """
-    if not tools or not model_reasoning_enabled_by_default(model_id):
+    if (
+        not tools and not skills_configuration
+    ) or not model_reasoning_enabled_by_default(model_id):
         return reasoning
 
     base = reasoning or OpenAIResponseReasoning()

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/benchmarks/data/python_1000_lines.py

@@ -922,7 +922,7 @@ print_rag_response(response)
 # - tisk odpovědi
 
 
-llm = ChatOpenAI(model_name="gpt-4o-mini")
+llm = ChatOpenAI(model_name="gpt-5.6-luna")
 response = llm.invoke("Say Hi")
 print(response)
 
@@ -933,7 +933,7 @@ print(response)
 # - tisk odpovědi v naformátované podobě
 
 
-llm = ChatOpenAI(model_name="gpt-4o-mini")
+llm = ChatOpenAI(model_name="gpt-5.6-luna")
 response = llm.invoke("Say Hi")
 print(response)
 
@@ -944,7 +944,7 @@ print(response)
 # - tisk odpovědi
 
 
-llm = ChatOpenAI(model_name="gpt-4o-mini", temperature=0.7)
+llm = ChatOpenAI(model_name="gpt-5.6-luna", temperature=0.7)
 response = llm.invoke("Say Hi")
 print(response)
 
@@ -957,7 +957,7 @@ print(response)
 
 prompt = PromptTemplate.from_template("How to say 'hi' in {language} language?")
 
-llm = ChatOpenAI(model_name="gpt-4o-mini")
+llm = ChatOpenAI(model_name="gpt-5.6-luna")
 chain = LLMChain(prompt=prompt, llm=llm)
 
 response = chain.invoke("Czech")
@@ -976,7 +976,7 @@ prompt = PromptTemplate(
 )
 
 
-llm = ChatOpenAI(model_name="gpt-4o-mini")
+llm = ChatOpenAI(model_name="gpt-5.6-luna")
 chain = LLMChain(prompt=prompt, llm=llm)
 
 response = chain.invoke("Czech")
@@ -990,7 +990,7 @@ print(response)
 
 prompt = PromptTemplate.from_template("How to say 'hi' in {language} language?")
 
-llm = ChatOpenAI(model_name="gpt-4o-mini")
+llm = ChatOpenAI(model_name="gpt-5.6-luna")
 
 chain = prompt | llm | StrOutputParser()
 

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/benchmarks/data_generators.py

@@ -43,7 +43,7 @@ def generate_model_for_provider(provider: str) -> str:
             "gpt-5.2 pro",
             "gpt-5 mini",
             "gpt-4.1",
-            "gpt-4o",
+            "gpt-5.6-luna",
             "gpt-4-turbo",
             "gpt-4.1 mini",
             "gpt-4.1 nano",

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/configuration/unified-mode/lightspeed-stack-invalid-config-and-legacy.yaml

@@ -22,4 +22,4 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/configuration/unified-mode/lightspeed-stack-invalid-providers-and-legacy.yaml

@@ -20,7 +20,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
   # Unified synthesis input (Decision S5): the high-level provider entry
   # replaces the default baseline's openai provider by id at synthesis time.
   providers:
@@ -28,4 +28,4 @@ inference:
       id: openai
       api_key_env: OPENAI_API_KEY
       allowed_models:
-        - ${env.E2E_OPENAI_MODEL:=gpt-4o-mini}
+        - ${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/configuration/unified-mode/lightspeed-stack-invalid-version-legacy-unified-body.yaml

@@ -18,7 +18,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
   # Unified synthesis input (Decision S5): the high-level provider entry
   # replaces the default baseline's openai provider by id at synthesis time.
   providers:
@@ -26,6 +26,6 @@ inference:
       id: openai
       api_key_env: OPENAI_API_KEY
       allowed_models:
-        - ${env.E2E_OPENAI_MODEL:=gpt-4o-mini}
+        - ${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}
 # INVALID: explicit legacy marker on a unified-shaped body (R11, LCORE-2872)
 config_format_version: legacy

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/configuration/unified-mode/lightspeed-stack-legacy-for-migration.yaml

@@ -20,4 +20,4 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-azure.yaml

@@ -19,13 +19,13 @@ providers:
     config: 
       base_url: https://ols-test.openai.azure.com/openai/v1
       api_version: 2024-02-15-preview
-      allowed_models: ["gpt-4o-mini"]
+      allowed_models: ["gpt-5.6-luna"]
       model_validation: false
   - provider_id: openai
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers
@@ -99,10 +99,10 @@ storage:
       backend: sql_default
 registered_resources:
   models:
-  - model_id: gpt-4o-mini
+  - model_id: gpt-5.6-luna
     provider_id: azure
     model_type: llm
-    provider_model_id: gpt-4o-mini
+    provider_model_id: gpt-5.6-luna
   - model_id: all-mpnet-base-v2
     model_type: embedding
     provider_id: sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-bedrock.yaml

@@ -23,7 +23,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-ci.yaml

@@ -18,7 +18,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-rhaiis.yaml

@@ -25,7 +25,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-rhelai.yaml

@@ -25,7 +25,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["gpt-4o-mini"]
+      allowed_models: ["gpt-5.6-luna"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-vertexai.yaml

@@ -24,7 +24,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configs/run-watsonx.yaml

@@ -25,7 +25,7 @@ providers:
     provider_type: remote::openai
     config:
       api_key: ${env.OPENAI_API_KEY}
-      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-4o-mini}"]
+      allowed_models: ["${env.E2E_OPENAI_MODEL:=gpt-5.6-luna}"]
   - config: {}
     provider_id: sentence-transformers
     provider_type: inline::sentence-transformers

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-authorized.yaml

@@ -36,7 +36,7 @@ authentication:
   module: "noop-with-token"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
 
 rag:
   byok:

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-byok-pdf.yaml

@@ -30,7 +30,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
 
 # BYOK store built from a PDF by rag-content's `pdf` module (LCORE-2091).
 # The vector_db_id is hardcoded to the id baked into the committed fixture

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-compaction-disabled.yaml

@@ -22,7 +22,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
   # Compaction e2e (LCORE-1673): a deliberately small window for every
   # model the e2e workflows run against, so the third query of the
   # compaction scenarios crosses the trigger threshold. The real windows
@@ -31,8 +31,8 @@ inference:
   # var, and context_windows keys are not env-substituted, so they are
   # not listed and skip the token-based trigger.
   context_windows:
-    openai/gpt-4o-mini: 2000
-    azure/gpt-4o-mini: 2000
+    openai/gpt-5.6-luna: 2000
+    azure/gpt-5.6-luna: 2000
     google-vertex/publishers/google/models/gemini-2.5-flash: 2000
     watsonx/meta-llama/llama-3-3-70b-instruct: 2000
     aws-bedrock/deepseek.v3-v1:0: 2000

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-compaction.yaml

@@ -22,7 +22,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
   # Compaction e2e (LCORE-1673): a deliberately small window for every
   # model the e2e workflows run against, so the third query of the
   # compaction scenarios crosses the trigger threshold. The real windows
@@ -31,8 +31,8 @@ inference:
   # var, and context_windows keys are not env-substituted, so they are
   # not listed and skip the token-based trigger.
   context_windows:
-    openai/gpt-4o-mini: 2000
-    azure/gpt-4o-mini: 2000
+    openai/gpt-5.6-luna: 2000
+    azure/gpt-5.6-luna: 2000
     google-vertex/publishers/google/models/gemini-2.5-flash: 2000
     watsonx/meta-llama/llama-3-3-70b-instruct: 2000
     aws-bedrock/deepseek.v3-v1:0: 2000

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-default.yaml

@@ -32,7 +32,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
 
 rag:
   byok:

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-legacy.yaml

@@ -22,7 +22,7 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna
 rag:
   byok:
     stores:

{lightspeed_stack-0.7.1 → lightspeed_stack-0.7.2}/tests/e2e/configuration/library-mode/lightspeed-stack-shields-empty.yaml

@@ -25,4 +25,4 @@ authentication:
   module: "noop"
 inference:
   default_provider: openai
-  default_model: gpt-4o-mini
+  default_model: gpt-5.6-luna