{"apiVersion":"v1","evaluationSchemaVersion":"1.0.0","kind":"owner_reported_claim","record":{"id":"gemma-4-26b-a4b-it-gpqa-diamond-owner-reported","claimType":"owner_reported","modelId":"gemma-4-26b-a4b-it","modelName":"Gemma 4 26B A4B IT","owner":"Google DeepMind","artifact":{"repository":"google/gemma-4-26B-A4B-it","revision":"4d7ae4984b7db7de8f8457170b3f1a419ee76d52"},"artifactAssociation":"artifact_snapshot_associated","executionArtifactDigest":null,"benchmark":{"suite":"GPQA","version":null,"subset":"Diamond"},"metric":{"name":"accuracy","value":82.3,"unit":"percentage_points","scale":"0-100","direction":"higher_is_better"},"evaluation":{"mode":"thinking","reportedSettings":[{"name":"model_stage","value":"instruction-tuned"},{"name":"thinking_mode","value":true}]},"sourceRefs":[{"sourceId":"gemma-4-26b-a4b-it-pinned-model-card","locator":"Benchmark Results, Gemma 4 26B A4B column, GPQA Diamond row"},{"sourceId":"gemma-4-technical-report","locator":"Table 5; all models are in thinking mode unless explicitly stated"}],"comparisonEligible":false,"missingContext":["Prompt, shot count, sampling parameters, sample count, and scoring implementation are not reported.","Benchmark task revision and evaluation harness are not reported.","Context limit, maximum output length, runtime, and hardware are not reported.","The provider does not publish a digest of the tensor artifact used for the evaluation."]},"sources":[{"id":"gemma-4-26b-a4b-it-pinned-model-card","title":"Gemma 4 26B A4B IT pinned model card","publisher":"Google DeepMind","url":"https://huggingface.co/google/gemma-4-26B-A4B-it/blob/4d7ae4984b7db7de8f8457170b3f1a419ee76d52/README.md","retrievedAt":"2026-08-03","sourceType":"official_model_card","artifactSnapshot":{"repository":"google/gemma-4-26B-A4B-it","revision":"4d7ae4984b7db7de8f8457170b3f1a419ee76d52"}},{"id":"gemma-4-technical-report","title":"Gemma 4 Technical Report","publisher":"Gemma Team, Google DeepMind","url":"https://arxiv.org/pdf/2607.02770v1","retrievedAt":"2026-08-03","sourceType":"technical_report"}],"rawArtifacts":[]}