Datasets:

WaltonFuture
/

vscode-bench-results

model stringclasses 15 values	model_name stringclasses 15 values	view stringclasses 2 values	benchmark stringclasses 10 values	metric stringclasses 35 values	score_0_100 float64 0 100	scored_cases int64 29 60	fixed_cases int64 31 60	metric_cases int64 2 60	note stringclasses 2 values
qwen21	Qwen21	original_scored_subset	geneval2	SoftTIFA_GM	13.041276	60	60	60
qwen21	Qwen21	original_scored_subset	geneval2	SoftTIFA_AM	56.938101	60	60	60
qwen21	Qwen21	original_scored_subset	genai_bench	VQAScore_all	80.227696	60	60	59
qwen21	Qwen21	original_scored_subset	genai_bench	VQAScore_basic	89.560077	60	60	25
qwen21	Qwen21	original_scored_subset	genai_bench	VQAScore_advanced	73.365652	60	60	34
qwen21	Qwen21	original_scored_subset	arena_t2i_hard	score	76.852294	31	31	31
qwen21	Qwen21	original_scored_subset	qwen_image_bench	QJudger_total	55.215497	60	60	60
qwen21	Qwen21	original_scored_subset	genexam	strict_score	0	60	60	60
qwen21	Qwen21	original_scored_subset	genexam	relaxed_score	46.071667	60	60	60
qwen21	Qwen21	original_scored_subset	grade	relax_score	56.2	52	52	52
qwen21	Qwen21	original_scored_subset	grade	acc_score	7.7	52	52	52
qwen21	Qwen21	original_scored_subset	risebench	Overall_normalized	46.180556	36	36	36
qwen21	Qwen21	original_scored_subset	paintbench	cie76_mean.mean_edit_accuracy	18.152368	60	60	57
qwen21	Qwen21	original_scored_subset	paintbench	cie76_mean.mean_preservation_accuracy	60.613772	60	60	57
qwen21	Qwen21	original_scored_subset	paintbench	cie76_mean.mean_iou	6.844248	60	60	57
qwen21	Qwen21	original_scored_subset	mmgr_image	arc.valid_solution	0	60	60	16
qwen21	Qwen21	original_scored_subset	mmgr_image	math.FINAL_CORRECT	0	60	60	11
qwen21	Qwen21	original_scored_subset	mmgr_image	maze.overall_score	7.686801	60	60	12
qwen21	Qwen21	original_scored_subset	mmgr_image	sudoku.overall_score	0	60	60	10
qwen21	Qwen21	original_scored_subset	mmgr_image	3D_real-world_navigation.overall_success	50	60	60	2
qwen21	Qwen21	original_scored_subset	mmgr_image	panoramic_view_last-mile_navigation.overall_success	0	60	60	4
qwen21	Qwen21	original_scored_subset	mmgr_image	simultaneous_localization_and_generation.overall_success	0	60	60	2
qwen21	Qwen21	original_scored_subset	mmgr_image	top-down_view_real-world_navigation.overall_success	0	60	60	3
qwen21	Qwen21	original_scored_subset	genpuzzle	figure_completion.official_normalized_score	13.8889	60	60	12
qwen21	Qwen21	original_scored_subset	genpuzzle	spatial_generation.official_normalized_score	0	60	60	2
qwen21	Qwen21	original_scored_subset	genpuzzle	maze_beginner.official_normalized_score	66.6667	60	60	2
qwen21	Qwen21	original_scored_subset	genpuzzle	maze_intermediate.official_normalized_score	66.6667	60	60	2
qwen21	Qwen21	original_scored_subset	genpuzzle	maze_advanced.official_normalized_score	33.3333	60	60	2
qwen21	Qwen21	original_scored_subset	genpuzzle	sudoku_reasoning.official_normalized_score	0	60	60	2
qwen21	Qwen21	original_scored_subset	genpuzzle	nonogram_reasoning.official_normalized_score	0	60	60	4
qwen21	Qwen21	original_scored_subset	genpuzzle	tangram_reasoning.official_normalized_score	8.3333	60	60	4
qwen21	Qwen21	original_scored_subset	genpuzzle	board_game_reasoning.descriptive_mean	34.777778	60	60	9	descriptive mean, not official aggregate
qwen21	Qwen21	original_scored_subset	genpuzzle	matchstick_reasoning.descriptive_mean	0	60	60	9	descriptive mean, not official aggregate
qwen21	Qwen21	original_scored_subset	genpuzzle	orthographic_reasoning.descriptive_mean	21.333333	60	60	3	descriptive mean, not official aggregate
qwen21	Qwen21	original_scored_subset	genpuzzle	math_visual_reasoning.descriptive_mean	28.888889	60	60	9	descriptive mean, not official aggregate
qwen21	Qwen21	common_15_models	geneval2	SoftTIFA_GM	13.041276	60	60	60
qwen21	Qwen21	common_15_models	geneval2	SoftTIFA_AM	56.938101	60	60	60
qwen21	Qwen21	common_15_models	genai_bench	VQAScore_all	80.227696	60	60	59
qwen21	Qwen21	common_15_models	genai_bench	VQAScore_basic	89.560077	60	60	25
qwen21	Qwen21	common_15_models	genai_bench	VQAScore_advanced	73.365652	60	60	34
qwen21	Qwen21	common_15_models	arena_t2i_hard	score	80.357354	29	31	29
qwen21	Qwen21	common_15_models	qwen_image_bench	QJudger_total	55.346268	59	60	59
qwen21	Qwen21	common_15_models	genexam	strict_score	0	59	60	59
qwen21	Qwen21	common_15_models	genexam	relaxed_score	46.411864	59	60	59
qwen21	Qwen21	common_15_models	grade	relax_score	56.2	52	52	52
qwen21	Qwen21	common_15_models	grade	acc_score	7.7	52	52	52
qwen21	Qwen21	common_15_models	risebench	Overall_normalized	45.357143	35	36	35
qwen21	Qwen21	common_15_models	paintbench	cie76_mean.mean_edit_accuracy	18.152368	60	60	57
qwen21	Qwen21	common_15_models	paintbench	cie76_mean.mean_preservation_accuracy	60.613772	60	60	57
qwen21	Qwen21	common_15_models	paintbench	cie76_mean.mean_iou	6.844248	60	60	57
qwen21	Qwen21	common_15_models	mmgr_image	arc.valid_solution	0	60	60	16
qwen21	Qwen21	common_15_models	mmgr_image	math.FINAL_CORRECT	0	60	60	11
qwen21	Qwen21	common_15_models	mmgr_image	maze.overall_score	7.686801	60	60	12
qwen21	Qwen21	common_15_models	mmgr_image	sudoku.overall_score	0	60	60	10
qwen21	Qwen21	common_15_models	mmgr_image	3D_real-world_navigation.overall_success	50	60	60	2
qwen21	Qwen21	common_15_models	mmgr_image	panoramic_view_last-mile_navigation.overall_success	0	60	60	4
qwen21	Qwen21	common_15_models	mmgr_image	simultaneous_localization_and_generation.overall_success	0	60	60	2
qwen21	Qwen21	common_15_models	mmgr_image	top-down_view_real-world_navigation.overall_success	0	60	60	3
qwen21	Qwen21	common_15_models	genpuzzle	figure_completion.official_normalized_score	13.8889	60	60	12
qwen21	Qwen21	common_15_models	genpuzzle	spatial_generation.official_normalized_score	0	60	60	2
qwen21	Qwen21	common_15_models	genpuzzle	maze_beginner.official_normalized_score	66.6667	60	60	2
qwen21	Qwen21	common_15_models	genpuzzle	maze_intermediate.official_normalized_score	66.6667	60	60	2
qwen21	Qwen21	common_15_models	genpuzzle	maze_advanced.official_normalized_score	33.3333	60	60	2
qwen21	Qwen21	common_15_models	genpuzzle	sudoku_reasoning.official_normalized_score	0	60	60	2
qwen21	Qwen21	common_15_models	genpuzzle	nonogram_reasoning.official_normalized_score	0	60	60	4
qwen21	Qwen21	common_15_models	genpuzzle	tangram_reasoning.official_normalized_score	8.3333	60	60	4
qwen21	Qwen21	common_15_models	genpuzzle	board_game_reasoning.descriptive_mean	34.777778	60	60	9	descriptive mean, not official aggregate
qwen21	Qwen21	common_15_models	genpuzzle	matchstick_reasoning.descriptive_mean	0	60	60	9	descriptive mean, not official aggregate
qwen21	Qwen21	common_15_models	genpuzzle	orthographic_reasoning.descriptive_mean	21.333333	60	60	3	descriptive mean, not official aggregate
qwen21	Qwen21	common_15_models	genpuzzle	math_visual_reasoning.descriptive_mean	28.888889	60	60	9	descriptive mean, not official aggregate
qwen21-turbo	Qwen21-Turbo	original_scored_subset	geneval2	SoftTIFA_GM	7.595558	60	60	60
qwen21-turbo	Qwen21-Turbo	original_scored_subset	geneval2	SoftTIFA_AM	52.242479	60	60	60
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genai_bench	VQAScore_all	81.215897	60	60	59
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genai_bench	VQAScore_basic	91.093227	60	60	25
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genai_bench	VQAScore_advanced	73.953155	60	60	34
qwen21-turbo	Qwen21-Turbo	original_scored_subset	arena_t2i_hard	score	75.504045	31	31	31
qwen21-turbo	Qwen21-Turbo	original_scored_subset	qwen_image_bench	QJudger_total	54.92634	60	60	60
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genexam	strict_score	0	60	60	60
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genexam	relaxed_score	45.938333	60	60	60
qwen21-turbo	Qwen21-Turbo	original_scored_subset	grade	relax_score	55.2	52	52	52
qwen21-turbo	Qwen21-Turbo	original_scored_subset	grade	acc_score	11.5	52	52	52
qwen21-turbo	Qwen21-Turbo	original_scored_subset	risebench	Overall_normalized	42.013889	36	36	36
qwen21-turbo	Qwen21-Turbo	original_scored_subset	paintbench	cie76_mean.mean_edit_accuracy	16.44042	60	60	57
qwen21-turbo	Qwen21-Turbo	original_scored_subset	paintbench	cie76_mean.mean_preservation_accuracy	55.724173	60	60	57
qwen21-turbo	Qwen21-Turbo	original_scored_subset	paintbench	cie76_mean.mean_iou	5.805017	60	60	57
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	arc.valid_solution	0	60	60	16
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	math.FINAL_CORRECT	0	60	60	11
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	maze.overall_score	14.491642	60	60	12
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	sudoku.overall_score	0	60	60	10
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	3D_real-world_navigation.overall_success	50	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	panoramic_view_last-mile_navigation.overall_success	0	60	60	4
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	simultaneous_localization_and_generation.overall_success	0	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	mmgr_image	top-down_view_real-world_navigation.overall_success	0	60	60	3
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	figure_completion.official_normalized_score	16.6667	60	60	12
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	spatial_generation.official_normalized_score	0	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	maze_beginner.official_normalized_score	33.3333	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	maze_intermediate.official_normalized_score	33.3333	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	maze_advanced.official_normalized_score	33.3333	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	sudoku_reasoning.official_normalized_score	0	60	60	2
qwen21-turbo	Qwen21-Turbo	original_scored_subset	genpuzzle	nonogram_reasoning.official_normalized_score	8.3333	60	60	4

End of preview. Expand in Data Studio

VSCode Bench Results

图片生成与编辑 / Image generation and editing

14 组本地原生图像模型 + 1 组 coding agent（gpt-6.1-sol / high），10 套 benchmark，每组固定 539 道图片题。

共 8,074 条有效评分、9 条内容过滤跳过、2 条源答案缺失；跳过和缺失不补零。

共同题目对比统一使用全部 15 组都有有效评分的 534 道题。各指标按 0–100 展示，保留原评分规则，不计算跨 benchmark 总分。上方数据查看器的 image_metrics 配置提供指标表，image_scores 提供逐题数据。

Downloads last month: 1,194