Evidence

Evidence connects prepared model facts to original observations. List a model’s records, aliases, receipts, source summaries, identity merges or unresolved conflicts, then request the original record explicitly.

Evidence pages retain source and receipt identifiers. Full original payloads are intentionally absent from model cards and metadata pages; fetch the record or checksum-addressed original document when needed. source filtering on model evidence is supported only for kind=records.

Original metadata, benchmark artifacts and history preserve their observation context. Links describe their destination separately from the source that discovered them. Merging model identities does not rewrite upstream receipts or invent official identifiers.

On this page

List model observation history

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

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
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.
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/history?limit=1'

Responses

StatusMeaningBody type
200Source-backed JSON responseObservationHistory
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.

{
    "data": [
        {
            "source_id": "openRouter",
            "source_record_id": "google/gemini-2.5-flash",
            "canonical_model_id": "google/gemini-2.5-flash",
            "revision_id": 1
        }
    ],
    "total_count": 1,
    "limit": 1,
    "offset": 0
}
FieldTypeRequiredMeaning / constraints
dataarray of objectYes
total_countintegerYes
limitintegerYes
offsetintegerYes

Read published original source metadata

GET /api/v1/sources/{source}/metadata/{checksum}

Exact UTF-8 metadata bytes returned as inert JSON, with SHA-256, media type, original URL, retrieval time and base64 body. HTML is returned as source text without execution. Content is visible only after complete receipt publication.

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
sourcepathYesstringRegistered data-source ID, such as openRouter. This identifies the collecting source, not necessarily the model developer.
checksumpathYesstringSHA-256 checksum identifying the immutable original document. Copy it from source evidence; do not guess it.

Request example

curl 'https://tensortrace.net/api/v1/sources/openRouter/metadata/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'

Responses

StatusMeaningBody type
200Published original metadataobject
400Invalid SHA-256No declared JSON body
404Unknown published originalNo declared JSON body

Response example · 200

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

{
    "source_id": "openRouter",
    "checksum_sha256": "example",
    "source_url": "example",
    "media_type": "example",
    "text": "example",
    "body_base64": "example"
}
FieldTypeRequiredMeaning / constraints
source_idstringYes
checksum_sha256stringYes
source_urlstringYes
media_typestringYes
textstringYes
body_base64stringYes

Immutable original benchmark export

GET /api/v1/sources/{source}/artifacts/{checksum}

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
sourcepathYesstringRegistered data-source ID, such as openRouter. This identifies the collecting source, not necessarily the model developer.
checksumpathYesstringSHA-256 checksum identifying the immutable original document. Copy it from source evidence; do not guess it.

Request example

curl 'https://tensortrace.net/api/v1/sources/openRouter/artifacts/aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa'

Responses

StatusMeaningBody type
200Original decoded UTF-8 export bytes, verified by SHA-256string
404Artifact is not linked to this sourceNo declared JSON body

Response example · 200

The response contains original bytes or an image. Content types: text/plain. Preserve the declared encoding and checksum; do not parse it as JSON.

Paginated source metadata; original payloads are separate evidence documents

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

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.
kindqueryNostring Default: "records"; Values: aliases, records, receipts, sources, merges, conflictsEvidence collection: records, aliases, receipts, sources, merges or conflicts. Default records.
limitqueryNointeger Default: 50; Minimum: 1; Maximum: 100Maximum 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.
sourcequeryNostringRegistered source ID. This filter is accepted only with kind=records.

Request example

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

Responses

StatusMeaningBody type
200Current local observationsDetailSectionPage
400Invalid requestNo declared JSON body
404Unknown model or sectionNo declared JSON body

Response example · 200

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

{
    "data": [
        {}
    ],
    "total_count": 1,
    "limit": 1,
    "offset": 0
}
FieldTypeRequiredMeaning / constraints
dataarray of objectYesMax. items: 100
total_countintegerYesMinimum: 0
limitintegerYesMinimum: 1; Maximum: 100
offsetintegerYesMinimum: 0

One complete original source observation

GET /api/v1/evidence/record

Access: public read; no TensorTrace API key.

Parameters

NameLocationRequiredType / limitsMeaning
sourcequeryYesstringRegistered data-source ID, such as openRouter. This identifies the collecting source, not necessarily the model developer.
recordqueryYesstringExact source_record_id from an evidence index entry. This is a source record key, not a canonical model ID.

Request example

curl 'https://tensortrace.net/api/v1/evidence/record?source=example&record=example'

Responses

StatusMeaningBody type
200Current local observationsSourceRecordDto
400Invalid requestNo declared JSON body
404Unknown model or sectionNo declared JSON body

Response example · 200

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

{
    "source_id": "openRouter",
    "source_record_id": "google/gemini-2.5-flash",
    "confidence": 1,
    "updated_at": "2026-10-05T00:00:00Z",
    "raw_payload": "example"
}
FieldTypeRequiredMeaning / constraints
source_idstringYes
source_record_idstringYes
receipt_idstring · uuid or nullNo
confidencenumberYes
updated_atstring · date-timeYes
raw_payloadany JSONYes
link_correctionsarray of SourceLinkCorrectionNo

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