SearchGen-Bench / data /manifest.json
wufangtai
Adopt ten-component benchmark protocol
a8782a2
Raw
History Blame Contribute Delete
3.27 kB
{
"artifacts": [
"prompt_scores.jsonl",
"leaderboard_overall.json",
"leaderboard_by_stratum.json",
"leaderboard_by_domain.json",
"leaderboard_by_failure_mode.json"
],
"benchmark": "SearchGen-Bench",
"dataset": {
"filename": "eval_datasets_548_finalized_v1.jsonl",
"n_prompts": 751,
"sha256": "2a80f209304421056b78d1fa2e8c0b24b64c33735cb18eee0044f17df30e13e7"
},
"generated_at": "2026-07-16T00:00:00+00:00",
"models": {
"bagel": {
"display_name": "Bagel",
"source_id": "bagel",
"type": "Open"
},
"gemini2d5flash": {
"display_name": "Nano Banana",
"source_id": "gemini2d5flash",
"type": "Commercial"
},
"gemini3pro": {
"display_name": "Nano Banana Pro",
"source_id": "gemini3pro",
"type": "Commercial"
},
"gpt_image": {
"display_name": "GPT-Image-2",
"source_id": "gpt_image",
"type": "Commercial"
},
"imagen3fast": {
"display_name": "Imagen3-Fast",
"source_id": "imagen3fast",
"type": "Commercial"
},
"jimeng4d0": {
"display_name": "SeedDream-4.0",
"source_id": "jimeng4d0",
"type": "Commercial"
},
"klein": {
"display_name": "Flux.2-Klein-9B",
"source_id": "klein",
"type": "Open"
},
"klein4b": {
"display_name": "Flux.2-Klein-4B",
"source_id": "klein4b",
"type": "Open"
},
"qwen1": {
"display_name": "Qwen-Image",
"source_id": "qwen1",
"type": "Open"
},
"qwen2": {
"display_name": "Qwen-Image-2",
"source_id": "qwen-image-2.0",
"type": "Commercial"
},
"qwen_image_2_pro": {
"display_name": "Qwen-Image-2-Pro",
"source_id": "qwen_image_2_pro",
"type": "Commercial"
},
"seedream4d5": {
"display_name": "SeedDream-4.5",
"source_id": "seedream4d5",
"type": "Commercial"
},
"xai_image": {
"display_name": "Grok-Imagine-Image",
"source_id": "xai_image",
"type": "Commercial"
}
},
"partition": {
"NoSearch": 100,
"SearchIntensive": 651,
"TextualSearch": 264,
"VisualSearch": 387
},
"schema_version": "1.0.0",
"scoring": {
"missing_policy_default": "zero_fill",
"missing_policy_exceptions": [
"qwen_image_2_pro",
"xai_image"
],
"overall_10_components": [
"checklist",
"rubric_adaptive",
"prompt_faithfulness",
"image_quality",
"text_rendering",
"ai_naturalness",
"composition_and_aesthetics",
"physical_plausibility",
"visual_reference_evaluation",
"text_reference_evaluation"
],
"overall_9_components": [
"checklist",
"rubric_adaptive",
"prompt_faithfulness",
"image_quality",
"text_rendering",
"ai_naturalness",
"composition_and_aesthetics",
"physical_plausibility",
"visual_reference_evaluation"
],
"overall_9_excluded_component": "text_reference_evaluation",
"primary_metric": "overall_10",
"public_rounding_decimals": 1,
"scale_public": [
0,
100
],
"scale_raw": [
0,
3
],
"source": "augmented_parsed_result_ffjudge_pp.json"
},
"source_commit": null
}