Operational Metrics

Operational metrics describe provider routes and workloads, not intrinsic model intelligence. Endpoint observations retain the OpenRouter source identifier, retrieval time and a stale flag. Missing or unavailable observations remain explicit; reads do not probe providers.

Normalized throughput, latency and time-to-first-token values retain percentile and workload context where available. Compare identical scopes, not an aggregate with a single region or provider. Listed price history is a series of collected quotes; it does not represent customers’ effective spending or a continuous upstream time series.

On this page

Read provider endpoint observations

GET /api/v1/models/{provider}/{model}/endpoints

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
providerpathYesstringCanonical model publisher namespace; URL-encode this path segment.
modelpathYesstringCanonical model name within the publisher namespace. URL-encode nested slashes in this single segment.

Request example

curl 'https://tensortrace.net/api/v1/models/google/gemini-2.5-flash/endpoints'

Responses

StatusMeaningBody type
200Source-backed JSON responseobject
400Invalid queryNo declared JSON body
404Unknown model/sourceNo declared JSON body
500Database unavailableNo declared JSON body
503Service temporarily unavailableNo declared JSON body

Response example · 200

Illustrative values; schema-valid and not a live catalog measurement.

{}

Type: object.

Provider-route operational snapshot with normalized units

GET /api/v1/models/{provider}/{model}/metrics

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
providerpathYesstringCanonical model publisher namespace; URL-encode this path segment.
modelpathYesstringCanonical model name within the publisher namespace. URL-encode nested slashes in this single segment.

Request example

curl 'https://tensortrace.net/api/v1/models/google/gemini-2.5-flash/metrics'

Responses

StatusMeaningBody type
200Source-backed JSON responseModelMetricsResponse
400Invalid queryNo declared JSON body
404Unknown model/sourceNo declared JSON body
500Database unavailableNo declared JSON body
503Service temporarily unavailableNo declared JSON body

Response example · 200

Illustrative values; schema-valid and not a live catalog measurement.

{
    "canonical_model_id": "google/gemini-2.5-flash",
    "source_id": "openRouter",
    "source_model_id": "google/gemini-2.5-flash",
    "source_url": "example",
    "retrieved_at": "2026-10-05T00:00:00Z",
    "stale": false,
    "coverage": "listed",
    "window": "30m",
    "providers": [
        {
            "provider_name": "example",
            "route": "example",
            "endpoint_name": "example",
            "status": 1,
            "latency_seconds": {
                "p50": 1,
                "p75": 1,
                "p90": 1,
                "p99": 1
            },
            "latency_kind": "time_to_first_token",
            "throughput_tokens_per_second": {
                "p50": 1,
                "p75": 1,
                "p90": 1,
                "p99": 1
            },
            "uptime_percent": {
                "last_5m": 1,
                "last_30m": 1,
                "last_1d": 1
            },
            "workloads": {}
        }
    ]
}
FieldTypeRequiredMeaning / constraints
canonical_model_idstringYes
source_idany JSONYes
source_model_idstring,nullYes
source_urlstring,nullYes
retrieved_atstring,null · date-timeYes
stalebooleanYes
coverageany JSONYesValues: listed, not_listed
windowany JSONYes
providersarray of EndpointMetricYes

Observed listed token quotes from immutable source history

GET /api/v1/models/{provider}/{model}/price-history

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
providerpathYesstringCanonical model publisher namespace; URL-encode this path segment.
modelpathYesstringCanonical model name within the publisher namespace. URL-encode nested slashes in this single segment.
limitqueryNointeger Default: 500; Minimum: 1; Maximum: 1000Maximum number of returned rows. The documented default and range apply; numeric pagination limits are clamped unless stated otherwise.
offsetqueryNointeger Default: 0; Minimum: 0Zero-based number of matching rows to skip. Default 0; negative values are clamped to 0.

Request example

curl 'https://tensortrace.net/api/v1/models/google/gemini-2.5-flash/price-history?limit=1'

Responses

StatusMeaningBody type
200Source-backed JSON responseModelPriceHistoryResponse
400Invalid queryNo declared JSON body
404Unknown model/sourceNo declared JSON body
500Database unavailableNo declared JSON body
503Service temporarily unavailableNo declared JSON body

Response example · 200

Illustrative values; schema-valid and not a live catalog measurement.

{
    "canonical_model_id": "google/gemini-2.5-flash",
    "source_id": "openRouter",
    "quote_kind": "listed_catalog_quote",
    "data": [
        {
            "source_model_id": "google/gemini-2.5-flash",
            "receipt_id": "00000000-0000-4000-8000-000000000001",
            "source_url": "example",
            "observed_at": "2026-10-05T00:00:00Z",
            "unit": "input_tokens_per_million",
            "amount_usd": 0.3,
            "currency": "USD",
            "raw_field": "example",
            "original_amount": "example",
            "source_pricing": {}
        }
    ],
    "total_observations": 1,
    "returned_observations": 1,
    "limit": 1,
    "offset": 0,
    "has_more": false
}
FieldTypeRequiredMeaning / constraints
canonical_model_idstringYes
source_idany JSONYes
quote_kindany JSONYes
dataarray of ListedQuoteObservationYes
total_observationsintegerYesMinimum: 0
returned_observationsintegerYesMinimum: 0
limitintegerYesMinimum: 1; Maximum: 1000
offsetintegerYesMinimum: 0
has_morebooleanYes

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace