generate openapi spec

docs/_static/llama-stack-spec.html

@@ -898,6 +898,49 @@
                         "schema": {
                             "type": "string"
                         }
+                    },
+                    {
+                        "name": "start_time",
+                        "in": "query",
+                        "required": true,
+                        "schema": {
+                            "type": "integer"
+                        }
+                    },
+                    {
+                        "name": "end_time",
+                        "in": "query",
+                        "required": false,
+                        "schema": {
+                            "type": "integer"
+                        }
+                    },
+                    {
+                        "name": "step",
+                        "in": "query",
+                        "required": false,
+                        "schema": {
+                            "type": "string"
+                        }
+                    },
+                    {
+                        "name": "query_type",
+                        "in": "query",
+                        "required": true,
+                        "schema": {
+                            "$ref": "#/components/schemas/MetricQueryType"
+                        }
+                    },
+                    {
+                        "name": "label_matchers",
+                        "in": "query",
+                        "required": false,
+                        "schema": {
+                            "type": "array",
+                            "items": {
+                                "$ref": "#/components/schemas/MetricLabelMatcher"
+                            }
+                        }
                     }
                 ]
             }
@@ -3559,6 +3602,12 @@
             "CompletionResponse": {
                 "type": "object",
                 "properties": {
+                    "metrics": {
+                        "type": "array",
+                        "items": {
+                            "$ref": "#/components/schemas/Metric"
+                        }
+                    },
                     "content": {
                         "type": "string",
                         "description": "The generated completion text"
@@ -3926,6 +3975,12 @@
             "CompletionResponseStreamChunk": {
                 "type": "object",
                 "properties": {
+                    "metrics": {
+                        "type": "array",
+                        "items": {
+                            "$ref": "#/components/schemas/Metric"
+                        }
+                    },
                     "delta": {
                         "type": "string",
                         "description": "New content generated since last chunk. This can be one or more tokens."
@@ -5092,6 +5147,12 @@
             "EmbeddingsResponse": {
                 "type": "object",
                 "properties": {
+                    "metrics": {
+                        "type": "array",
+                        "items": {
+                            "$ref": "#/components/schemas/Metric"
+                        }
+                    },
                     "embeddings": {
                         "type": "array",
                         "items": {

docs/_static/llama-stack-spec.yaml

@@ -2236,6 +2236,10 @@ components:
     CompletionResponse:
       type: object
       properties:
+        metrics:
+          type: array
+          items:
+            $ref: '#/components/schemas/Metric'
         content:
           type: string
           description: The generated completion text
@@ -2554,6 +2558,10 @@ components:
     CompletionResponseStreamChunk:
       type: object
       properties:
+        metrics:
+          type: array
+          items:
+            $ref: '#/components/schemas/Metric'
         delta:
           type: string
           description: >-
@@ -3341,6 +3349,10 @@ components:
     EmbeddingsResponse:
       type: object
       properties:
+        metrics:
+          type: array
+          items:
+            $ref: '#/components/schemas/Metric'
         embeddings:
           type: array
           items:

llama_stack/apis/telemetry/telemetry.py

@@ -235,15 +235,23 @@ class MetricsMixin(BaseModel):
 
 @json_schema_type
 class MetricQueryType(Enum):
-    RANGE = "range"  # Returns data points over time range
+    RANGE = "range"
-    INSTANT = "instant"  # Returns single data point
+    INSTANT = "instant"
+
+
+@json_schema_type
+class MetricLabelOperator(Enum):
+    EQUALS = "="
+    NOT_EQUALS = "!="
+    REGEX_MATCH = "=~"
+    REGEX_NOT_MATCH = "!~"
 
 
 @json_schema_type
 class MetricLabelMatcher(BaseModel):
     name: str
     value: str
-    operator: Literal["=", "!=", "=~", "!~"] = "="  # Prometheus-style operators
+    operator: MetricLabelOperator = MetricLabelOperator.EQUALS
 
 
 @json_schema_type
@@ -313,9 +321,9 @@ class Telemetry(Protocol):
     async def get_metrics(
         self,
         metric_name: str,
-        start_time: int,  # Unix timestamp in seconds
+        start_time: int,
-        end_time: Optional[int] = None,  # Unix timestamp in seconds
+        end_time: Optional[int] = None,
-        step: Optional[str] = "1m",  # Prometheus-style duration: 1m, 5m, 1h, etc.
+        step: Optional[str] = "1m",
         query_type: MetricQueryType = MetricQueryType.RANGE,
         label_matchers: Optional[List[MetricLabelMatcher]] = None,
     ) -> GetMetricsResponse: ...