{"apiVersion":"v1","evaluationSchemaVersion":"1.0.0","kind":"owner_reported_claim","record":{"id":"deepseek-v4-pro-gpqa-diamond-owner-reported","claimType":"owner_reported","modelId":"deepseek-v4-pro","modelName":"DeepSeek V4 Pro","owner":"DeepSeek","artifact":{"repository":"deepseek-ai/DeepSeek-V4-Pro","revision":"b5968e9190ef611bbf34a7229255be88a0e937c1"},"artifactAssociation":"artifact_snapshot_associated","executionArtifactDigest":null,"benchmark":{"suite":"GPQA","version":null,"subset":"Diamond"},"metric":{"name":"Pass@1","value":90.1,"unit":"percentage_points","scale":"0-100","direction":"higher_is_better"},"evaluation":{"mode":"think_max","reportedSettings":[{"name":"reasoning_mode","value":"Think Max"},{"name":"temperature","value":1},{"name":"context_tokens","value":384000},{"name":"system_prompt_variant","value":"Think Max special system prompt"}]},"sourceRefs":[{"sourceId":"deepseek-v4-pro-pinned-model-card","locator":"Evaluation Results, Comparison across Modes, GPQA Diamond row, V4-Pro Max column"},{"sourceId":"deepseek-v4-technical-report","locator":"Section 5.3.1 Evaluation Setup and Table 7"}],"comparisonEligible":false,"missingContext":["Top-p, exact prompt text for GPQA, shot count, and sample count are not reported.","Benchmark revision and evaluation harness version are not reported.","Inference engine and hardware are not reported.","The provider does not publish a digest of the tensor artifact used for the evaluation."]},"sources":[{"id":"deepseek-v4-pro-pinned-model-card","title":"DeepSeek V4 Pro pinned model card","publisher":"DeepSeek","url":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/b5968e9190ef611bbf34a7229255be88a0e937c1/README.md","retrievedAt":"2026-08-03","sourceType":"official_model_card","artifactSnapshot":{"repository":"deepseek-ai/DeepSeek-V4-Pro","revision":"b5968e9190ef611bbf34a7229255be88a0e937c1"}},{"id":"deepseek-v4-technical-report","title":"DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence","publisher":"DeepSeek-AI","url":"https://arxiv.org/pdf/2606.19348v1","retrievedAt":"2026-08-03","sourceType":"technical_report"}],"rawArtifacts":[]}