Spaces:
Running
Running
Yuxuan-Zhang-Dexter
commited on
Commit
·
8e60916
1
Parent(s):
b308c41
update latest leaderboard info
Browse files- app.py +14 -0
- assets/model_color.json +5 -1
- assets/news.json +6 -0
- data_visualization.py +19 -11
- rank_data_03_25_2025.json +191 -55
app.py
CHANGED
@@ -522,6 +522,14 @@ def build_app():
|
|
522 |
max-height: none !important;
|
523 |
position: relative !important;
|
524 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
525 |
|
526 |
/* Force scrolling to work on the main container */
|
527 |
.gradio-container, #root, #app {
|
@@ -708,6 +716,7 @@ def build_app():
|
|
708 |
|
709 |
# Add custom JavaScript for table header line breaks
|
710 |
gr.HTML("""
|
|
|
711 |
<script>
|
712 |
// Function to add line breaks to table headers
|
713 |
function formatTableHeaders() {
|
@@ -847,6 +856,7 @@ def build_app():
|
|
847 |
with gr.Column(visible=True) as overall_visualizations:
|
848 |
with gr.Tabs():
|
849 |
with gr.Tab("📈 Radar Chart"):
|
|
|
850 |
radar_visualization = gr.Plot(
|
851 |
label="Comparative Analysis (Radar Chart)",
|
852 |
elem_classes="visualization-container"
|
@@ -857,6 +867,10 @@ def build_app():
|
|
857 |
# label="Comparative Analysis (Group Bar Chart)",
|
858 |
# elem_classes="visualization-container"
|
859 |
# )
|
|
|
|
|
|
|
|
|
860 |
|
861 |
# Hidden placeholder for group bar visualization (to maintain code references)
|
862 |
group_bar_visualization = gr.Plot(visible=False)
|
|
|
522 |
max-height: none !important;
|
523 |
position: relative !important;
|
524 |
}
|
525 |
+
.radar-tip {
|
526 |
+
font-size: 14px;
|
527 |
+
color: #555;
|
528 |
+
margin-top: 5px;
|
529 |
+
margin-bottom: 20px;
|
530 |
+
font-style: italic;
|
531 |
+
}
|
532 |
+
|
533 |
|
534 |
/* Force scrolling to work on the main container */
|
535 |
.gradio-container, #root, #app {
|
|
|
716 |
|
717 |
# Add custom JavaScript for table header line breaks
|
718 |
gr.HTML("""
|
719 |
+
|
720 |
<script>
|
721 |
// Function to add line breaks to table headers
|
722 |
function formatTableHeaders() {
|
|
|
856 |
with gr.Column(visible=True) as overall_visualizations:
|
857 |
with gr.Tabs():
|
858 |
with gr.Tab("📈 Radar Chart"):
|
859 |
+
|
860 |
radar_visualization = gr.Plot(
|
861 |
label="Comparative Analysis (Radar Chart)",
|
862 |
elem_classes="visualization-container"
|
|
|
867 |
# label="Comparative Analysis (Group Bar Chart)",
|
868 |
# elem_classes="visualization-container"
|
869 |
# )
|
870 |
+
gr.Markdown(
|
871 |
+
"*💡 Click a legend entry to isolate that model. Double-click additional ones to add them for comparison.*",
|
872 |
+
elem_classes="radar-tip"
|
873 |
+
)
|
874 |
|
875 |
# Hidden placeholder for group bar visualization (to maintain code references)
|
876 |
group_bar_visualization = gr.Plot(visible=False)
|
assets/model_color.json
CHANGED
@@ -6,13 +6,17 @@
|
|
6 |
"gemini-2.0-flash": "#FF4081",
|
7 |
"gemini-2.0-flash-thinking-exp-1219": "#C2185B",
|
8 |
"gemini-2.5-pro-exp-03-25": "#FF80AB",
|
|
|
9 |
"gpt-4o-2024-11-20": "#00BFA5",
|
10 |
"gpt-4.5-preview-2025-02-27": "#00796B",
|
11 |
"gpt-4.1-2025-04-14": "#00897B",
|
12 |
"o1-2024-12-17": "#4DB6AC",
|
13 |
"o1-mini-2024-09-12": "#26A69A",
|
14 |
"o3-mini-2025-01-31(medium)": "#80CBC4",
|
|
|
|
|
|
|
15 |
"deepseek-v3": "#FFC107",
|
16 |
"deepseek-r1": "#FFA000",
|
17 |
-
"
|
18 |
}
|
|
|
6 |
"gemini-2.0-flash": "#FF4081",
|
7 |
"gemini-2.0-flash-thinking-exp-1219": "#C2185B",
|
8 |
"gemini-2.5-pro-exp-03-25": "#FF80AB",
|
9 |
+
"gemini-2.5-flash-preview-04-17": "#F06292",
|
10 |
"gpt-4o-2024-11-20": "#00BFA5",
|
11 |
"gpt-4.5-preview-2025-02-27": "#00796B",
|
12 |
"gpt-4.1-2025-04-14": "#00897B",
|
13 |
"o1-2024-12-17": "#4DB6AC",
|
14 |
"o1-mini-2024-09-12": "#26A69A",
|
15 |
"o3-mini-2025-01-31(medium)": "#80CBC4",
|
16 |
+
"o3": "#26C6DA",
|
17 |
+
"o4-mini": "#00ACC1",
|
18 |
+
"grok-3-beta": "#FF7043",
|
19 |
"deepseek-v3": "#FFC107",
|
20 |
"deepseek-r1": "#FFA000",
|
21 |
+
"llama-4-maverick-17b-128e-instruct-fp8": "#8E24AA"
|
22 |
}
|
assets/news.json
CHANGED
@@ -1,5 +1,11 @@
|
|
1 |
{
|
2 |
"news": [
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
{
|
4 |
"date": "2025-04-15",
|
5 |
"video_link": "https://www.youtube.com/watch?v=q8PMW870yp8",
|
|
|
1 |
{
|
2 |
"news": [
|
3 |
+
{
|
4 |
+
"date": "2025-04-23",
|
5 |
+
"video_link": "https://www.youtube.com/watch?v=NB1-5aKV9v4",
|
6 |
+
"twitter_text": "Zero-Shot AI Gaming Showdown: O3 Multi-Modal Might Sweeps Sokoban & 2048, Lands Top-2 in Phoenix Wright & Candy Crush",
|
7 |
+
"twitter_link": "https://x.com/haoailab"
|
8 |
+
},
|
9 |
{
|
10 |
"date": "2025-04-15",
|
11 |
"video_link": "https://www.youtube.com/watch?v=q8PMW870yp8",
|
data_visualization.py
CHANGED
@@ -338,16 +338,21 @@ def create_single_radar_chart(df, selected_games=None, highlight_models=None):
|
|
338 |
mean, std = vals.mean(), vals.std()
|
339 |
df[f"norm_{col}"] = normalize_values(vals, mean, std)
|
340 |
|
341 |
-
# Group players by prefix
|
342 |
model_groups = {}
|
343 |
for player in df["Player"]:
|
344 |
prefix = get_model_prefix(player)
|
345 |
model_groups.setdefault(prefix, []).append(player)
|
346 |
-
|
347 |
-
#
|
|
|
|
|
|
|
|
|
|
|
348 |
grouped_players = []
|
349 |
-
for prefix in
|
350 |
-
grouped_players.extend(
|
351 |
|
352 |
fig = go.Figure()
|
353 |
|
@@ -363,12 +368,15 @@ def create_single_radar_chart(df, selected_games=None, highlight_models=None):
|
|
363 |
|
364 |
r = [row[f"norm_{col}"] for col in game_cols]
|
365 |
|
|
|
|
|
|
|
366 |
fig.add_trace(go.Scatterpolar(
|
367 |
r=r + [r[0]],
|
368 |
theta=categories + [categories[0]],
|
369 |
mode='lines+markers',
|
370 |
fill='toself',
|
371 |
-
name=
|
372 |
line=dict(color=color, width=4 if is_highlighted else 2),
|
373 |
marker=dict(color=color),
|
374 |
fillcolor=fillcolor,
|
@@ -379,7 +387,7 @@ def create_single_radar_chart(df, selected_games=None, highlight_models=None):
|
|
379 |
fig.update_layout(
|
380 |
autosize=False,
|
381 |
width=1000,
|
382 |
-
height=
|
383 |
margin=dict(l=400, r=200, t=20, b=20),
|
384 |
title=dict(
|
385 |
text="AI Normalized Performance Across Games",
|
@@ -407,12 +415,12 @@ def create_single_radar_chart(df, selected_games=None, highlight_models=None):
|
|
407 |
),
|
408 |
legend=dict(
|
409 |
font=dict(size=12),
|
410 |
-
title="Choose your model
|
411 |
itemsizing='trace',
|
412 |
-
x=-1.
|
413 |
-
y=0.8,
|
414 |
-
xanchor='left',
|
415 |
yanchor='top',
|
|
|
416 |
bgcolor='rgba(255,255,255,0.6)',
|
417 |
bordercolor='gray',
|
418 |
borderwidth=1
|
|
|
338 |
mean, std = vals.mean(), vals.std()
|
339 |
df[f"norm_{col}"] = normalize_values(vals, mean, std)
|
340 |
|
341 |
+
# Group players by prefix and sort alphabetically
|
342 |
model_groups = {}
|
343 |
for player in df["Player"]:
|
344 |
prefix = get_model_prefix(player)
|
345 |
model_groups.setdefault(prefix, []).append(player)
|
346 |
+
|
347 |
+
# Sort each group alphabetically
|
348 |
+
for prefix in model_groups:
|
349 |
+
model_groups[prefix] = sorted(model_groups[prefix], key=str.lower)
|
350 |
+
|
351 |
+
# Get sorted prefixes and create ordered player list
|
352 |
+
sorted_prefixes = sorted(model_groups.keys(), key=str.lower)
|
353 |
grouped_players = []
|
354 |
+
for prefix in sorted_prefixes:
|
355 |
+
grouped_players.extend(model_groups[prefix])
|
356 |
|
357 |
fig = go.Figure()
|
358 |
|
|
|
368 |
|
369 |
r = [row[f"norm_{col}"] for col in game_cols]
|
370 |
|
371 |
+
# Convert player name to lowercase for the legend
|
372 |
+
display_name = player.lower()
|
373 |
+
|
374 |
fig.add_trace(go.Scatterpolar(
|
375 |
r=r + [r[0]],
|
376 |
theta=categories + [categories[0]],
|
377 |
mode='lines+markers',
|
378 |
fill='toself',
|
379 |
+
name=display_name, # Use lowercase name in legend
|
380 |
line=dict(color=color, width=4 if is_highlighted else 2),
|
381 |
marker=dict(color=color),
|
382 |
fillcolor=fillcolor,
|
|
|
387 |
fig.update_layout(
|
388 |
autosize=False,
|
389 |
width=1000,
|
390 |
+
height=620, # Increased height to accommodate legend
|
391 |
margin=dict(l=400, r=200, t=20, b=20),
|
392 |
title=dict(
|
393 |
text="AI Normalized Performance Across Games",
|
|
|
415 |
),
|
416 |
legend=dict(
|
417 |
font=dict(size=12),
|
418 |
+
title="Choose your model 💡 (click / double-click)",
|
419 |
itemsizing='trace',
|
420 |
+
x=-1.4, # Moved further left
|
421 |
+
y=0.8, # Moved to top
|
|
|
422 |
yanchor='top',
|
423 |
+
xanchor='left',
|
424 |
bgcolor='rgba(255,255,255,0.6)',
|
425 |
bordercolor='gray',
|
426 |
borderwidth=1
|
rank_data_03_25_2025.json
CHANGED
@@ -2,122 +2,157 @@
|
|
2 |
"Super Mario Bros": {
|
3 |
"runs": 5,
|
4 |
"results": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
{
|
6 |
"model": "claude-3-7-sonnet-20250219",
|
7 |
"score": 710,
|
8 |
"progress": "1-1",
|
9 |
"time_s": 64.2,
|
10 |
-
"rank":
|
11 |
},
|
12 |
{
|
13 |
"model": "gpt-4o-2024-11-20",
|
14 |
"score": 560,
|
15 |
"progress": "1-1",
|
16 |
"time_s": 58.6,
|
17 |
-
"rank":
|
18 |
},
|
19 |
{
|
20 |
"model": "gemini-2.0-flash",
|
21 |
"score": 320,
|
22 |
"progress": "1-1",
|
23 |
"time_s": 51.8,
|
24 |
-
"rank":
|
25 |
},
|
26 |
{
|
27 |
"model": "claude-3-5-haiku-20241022",
|
28 |
"score": 140,
|
29 |
"progress": "1-1",
|
30 |
"time_s": 76.4,
|
31 |
-
"rank":
|
32 |
},
|
33 |
{
|
34 |
"model": "gpt-4.5-preview-2025-02-27",
|
35 |
"score": 160,
|
36 |
"progress": "1-1",
|
37 |
"time_s": 62.8,
|
38 |
-
"rank":
|
39 |
}
|
40 |
]
|
41 |
},
|
42 |
"2048": {
|
43 |
"runs": 1,
|
44 |
"results": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
45 |
{
|
46 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
47 |
"score": 256,
|
48 |
"steps": 114,
|
49 |
"time": ">200",
|
50 |
-
"rank":
|
51 |
},
|
52 |
{
|
53 |
"model": "o1-2024-12-17",
|
54 |
"score": 256,
|
55 |
"steps": 116,
|
56 |
"time": ">200",
|
57 |
-
"rank":
|
58 |
},
|
59 |
{
|
60 |
"model": "claude-3-7-sonnet-20250219",
|
61 |
"score": 256,
|
62 |
"steps": 130,
|
63 |
"time": "20:36",
|
64 |
-
"rank":
|
65 |
},
|
66 |
{
|
67 |
"model": "deepseek-v3",
|
68 |
"score": 256,
|
69 |
"steps": 216,
|
70 |
"time": "54.02",
|
71 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
72 |
},
|
73 |
{
|
74 |
"model": "gemini-2.0-flash",
|
75 |
"score": 128,
|
76 |
"steps": 111,
|
77 |
"time": "18:43",
|
78 |
-
"rank":
|
79 |
},
|
80 |
{
|
81 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
82 |
"score": 128,
|
83 |
"steps": 132,
|
84 |
"time": ">100",
|
85 |
-
"rank":
|
86 |
},
|
87 |
{
|
88 |
"model": "gemini-2.5-pro-exp-03-25",
|
89 |
"score": 128,
|
90 |
"steps": 138,
|
91 |
"time": "169",
|
92 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
},
|
94 |
{
|
95 |
"model": "claude-3-5-sonnet-20241022",
|
96 |
"score": 64,
|
97 |
"steps": 92,
|
98 |
"time": "9:2",
|
99 |
-
"rank":
|
100 |
},
|
101 |
{
|
102 |
"model": "gpt-4.5-preview-2025-02-27",
|
103 |
"score": 34,
|
104 |
"steps": 34,
|
105 |
"time": "8:25",
|
106 |
-
"rank":
|
107 |
},
|
108 |
{
|
109 |
"model": "gpt-4o-2024-11-20",
|
110 |
"score": 16,
|
111 |
"steps": 21,
|
112 |
"time": "1:17",
|
113 |
-
"rank":
|
114 |
-
},
|
115 |
-
{
|
116 |
-
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
117 |
-
"score": 128,
|
118 |
-
"steps": 145,
|
119 |
-
"time": ">100",
|
120 |
-
"rank": 8
|
121 |
}
|
122 |
]
|
123 |
},
|
@@ -182,142 +217,207 @@
|
|
182 |
"Candy Crush": {
|
183 |
"runs": 3,
|
184 |
"results": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
185 |
{
|
186 |
"model": "o3-mini-2025-01-31(medium)",
|
187 |
"score_runs": "90;109;120",
|
188 |
"average_score": 106.33,
|
189 |
"steps": 25,
|
190 |
-
"rank":
|
191 |
},
|
192 |
{
|
193 |
"model": "o1-2024-12-17",
|
194 |
"score_runs": "96;114;83",
|
195 |
"average_score": 97.67,
|
196 |
"steps": 25,
|
197 |
-
"rank":
|
198 |
},
|
199 |
{
|
200 |
"model": "deepseek-r1",
|
201 |
"score_runs": "62;108;105",
|
202 |
"average_score": 91.67,
|
203 |
"steps": 25,
|
204 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
205 |
},
|
206 |
{
|
207 |
"model": "gemini-2.5-pro-exp-03-25",
|
208 |
"score_runs": "50;36;68",
|
209 |
"average_score": 51.33,
|
210 |
"steps": 25,
|
211 |
-
"rank":
|
212 |
},
|
213 |
{
|
214 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
215 |
"score_runs": "36;46;24",
|
216 |
"average_score": 35.33,
|
217 |
"steps": 25,
|
218 |
-
"rank":
|
219 |
},
|
220 |
{
|
221 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
222 |
"score_runs": "0;15;39",
|
223 |
"average_score": 18,
|
224 |
"steps": 25,
|
225 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
226 |
},
|
227 |
{
|
228 |
"model": "claude-3-5-sonnet-20241022",
|
229 |
"score_runs": "3;0;0",
|
230 |
"average_score": 1,
|
231 |
"steps": 25,
|
232 |
-
"rank":
|
233 |
},
|
234 |
{
|
235 |
"model": "deepseek-v3",
|
236 |
"score_runs": "0;0;0",
|
237 |
"average_score": 0,
|
238 |
"steps": 25,
|
239 |
-
"rank":
|
240 |
-
},
|
241 |
-
{
|
242 |
-
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
243 |
-
"score_runs": "6;0;0",
|
244 |
-
"average_score": 2,
|
245 |
-
"steps": 25,
|
246 |
-
"rank": 8
|
247 |
}
|
248 |
]
|
249 |
},
|
250 |
"Sokoban": {
|
251 |
"runs": 3,
|
252 |
"results": [
|
|
|
|
|
|
|
|
|
|
|
|
|
253 |
{
|
254 |
"model": "o3-mini-2025-01-31(medium)",
|
255 |
"levels_cracked": "2; 3; 2",
|
256 |
"steps": "[17,52,68];[24,58,78,91];[19,44,64]",
|
257 |
-
"rank":
|
258 |
},
|
259 |
{
|
260 |
"model": "gemini-2.5-pro-exp-03-25",
|
261 |
"levels_cracked": "2;2;3",
|
262 |
"steps": "[23, 46, 79]; [20,50,77]; [26,95,125,175]",
|
263 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
264 |
},
|
265 |
{
|
266 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
267 |
"levels_cracked": "1; 2; 0",
|
268 |
"steps": "[17,35];[15,40,43];[4]",
|
269 |
-
"rank":
|
270 |
},
|
271 |
{
|
272 |
"model": "o1-2024-12-17",
|
273 |
"levels_cracked": "1; 1; 1",
|
274 |
"steps": null,
|
275 |
-
"rank":
|
276 |
},
|
277 |
{
|
278 |
"model": "deepseek-r1",
|
279 |
"levels_cracked": "1; 0; 1",
|
280 |
"steps": "[19,42];[13];[19,36]",
|
281 |
"note": "stuck",
|
282 |
-
"rank":
|
283 |
},
|
284 |
{
|
285 |
"model": "o1-mini-2024-09-12",
|
286 |
"levels_cracked": "0;1;0",
|
287 |
"steps": null,
|
288 |
-
"rank":
|
289 |
},
|
290 |
{
|
291 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
292 |
"levels_cracked": "0; 0; 0",
|
293 |
"steps": "[23]; [14]; [14]",
|
294 |
-
"rank":
|
295 |
},
|
296 |
{
|
297 |
"model": "gpt-4o-2024-11-20",
|
298 |
"levels_cracked": "0; 0; 0",
|
299 |
"steps": "[68];[105];[168]",
|
300 |
"note": "stuck in a loop",
|
301 |
-
"rank":
|
302 |
},
|
303 |
{
|
304 |
"model": "claude-3-5-sonnet-20241022",
|
305 |
"levels_cracked": "0; 0; 0",
|
306 |
"steps": "[21]; [30]; [51]",
|
307 |
"note": "stuck in a loop",
|
308 |
-
"rank":
|
309 |
},
|
310 |
{
|
311 |
"model": "deepseek-v3",
|
312 |
"levels_cracked": "0; 0; 0",
|
313 |
"steps": "[9]; [47]; [64]",
|
314 |
-
"rank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
315 |
},
|
316 |
{
|
317 |
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
318 |
"levels_cracked": "0;0;0",
|
319 |
"steps": "[5]",
|
320 |
-
"rank":
|
321 |
}
|
322 |
]
|
323 |
},
|
@@ -333,12 +433,21 @@
|
|
333 |
"score": 26,
|
334 |
"note": "stuck at the end not present evidence"
|
335 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
336 |
{
|
337 |
"model": "gemini-2.5-pro-exp-03-25",
|
338 |
"levels_cracked": "2; 3",
|
339 |
"lives_left": "[5,5,0]; [5, 5, 4, 0]",
|
340 |
"cracked_details": "4: 0/8",
|
341 |
-
"rank":
|
342 |
"score": 20,
|
343 |
"note": "failed to present evidence"
|
344 |
},
|
@@ -347,7 +456,7 @@
|
|
347 |
"levels_cracked": "1; 1",
|
348 |
"lives_left": "[3,0]; [5,0]",
|
349 |
"cracked_details": "2: 3/9",
|
350 |
-
"rank":
|
351 |
"score": 8,
|
352 |
"note": "failed to present evidence"
|
353 |
},
|
@@ -356,7 +465,7 @@
|
|
356 |
"levels_cracked": "1",
|
357 |
"lives_left": "5, 5",
|
358 |
"cracked_details": "1:1/8",
|
359 |
-
"rank":
|
360 |
"score": 6,
|
361 |
"note": "stuck in loop"
|
362 |
},
|
@@ -365,16 +474,25 @@
|
|
365 |
"levels_cracked": "1",
|
366 |
"lives_left": "[4,5]",
|
367 |
"cracked_details": "1: 1/8",
|
368 |
-
"rank":
|
369 |
"score": 6,
|
370 |
"note": "stuck in loop"
|
371 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
372 |
{
|
373 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
374 |
"levels_cracked": "0",
|
375 |
"lives_left": "0",
|
376 |
"cracked_details": "1: 4/5",
|
377 |
-
"rank":
|
378 |
"score": 4,
|
379 |
"note": "stuck in the last option section"
|
380 |
},
|
@@ -383,16 +501,34 @@
|
|
383 |
"levels_cracked": "0",
|
384 |
"lives_left": "0",
|
385 |
"cracked_details": "1: 4/5",
|
386 |
-
"rank":
|
387 |
"score": 4,
|
388 |
"note": "stuck in the 3rd evidence present"
|
389 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
390 |
{
|
391 |
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
392 |
"levels_cracked": "0",
|
393 |
"lives_left": "0",
|
394 |
"cracked_details": "0:0/5",
|
395 |
-
"rank":
|
396 |
"score": 0,
|
397 |
"note": "failed to present evidence"
|
398 |
}
|
|
|
2 |
"Super Mario Bros": {
|
3 |
"runs": 5,
|
4 |
"results": [
|
5 |
+
{
|
6 |
+
"model": "gpt-4.1-2025-04-14",
|
7 |
+
"score": 740,
|
8 |
+
"progress": "1-1",
|
9 |
+
"time_s": 68.6,
|
10 |
+
"rank": 1
|
11 |
+
},
|
12 |
{
|
13 |
"model": "claude-3-7-sonnet-20250219",
|
14 |
"score": 710,
|
15 |
"progress": "1-1",
|
16 |
"time_s": 64.2,
|
17 |
+
"rank": 2
|
18 |
},
|
19 |
{
|
20 |
"model": "gpt-4o-2024-11-20",
|
21 |
"score": 560,
|
22 |
"progress": "1-1",
|
23 |
"time_s": 58.6,
|
24 |
+
"rank": 3
|
25 |
},
|
26 |
{
|
27 |
"model": "gemini-2.0-flash",
|
28 |
"score": 320,
|
29 |
"progress": "1-1",
|
30 |
"time_s": 51.8,
|
31 |
+
"rank": 4
|
32 |
},
|
33 |
{
|
34 |
"model": "claude-3-5-haiku-20241022",
|
35 |
"score": 140,
|
36 |
"progress": "1-1",
|
37 |
"time_s": 76.4,
|
38 |
+
"rank": 5
|
39 |
},
|
40 |
{
|
41 |
"model": "gpt-4.5-preview-2025-02-27",
|
42 |
"score": 160,
|
43 |
"progress": "1-1",
|
44 |
"time_s": 62.8,
|
45 |
+
"rank": 6
|
46 |
}
|
47 |
]
|
48 |
},
|
49 |
"2048": {
|
50 |
"runs": 1,
|
51 |
"results": [
|
52 |
+
{
|
53 |
+
"model": "o3",
|
54 |
+
"score": 256,
|
55 |
+
"steps": 108,
|
56 |
+
"time": "58:09",
|
57 |
+
"rank": 1
|
58 |
+
},
|
59 |
{
|
60 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
61 |
"score": 256,
|
62 |
"steps": 114,
|
63 |
"time": ">200",
|
64 |
+
"rank": 2
|
65 |
},
|
66 |
{
|
67 |
"model": "o1-2024-12-17",
|
68 |
"score": 256,
|
69 |
"steps": 116,
|
70 |
"time": ">200",
|
71 |
+
"rank": 3
|
72 |
},
|
73 |
{
|
74 |
"model": "claude-3-7-sonnet-20250219",
|
75 |
"score": 256,
|
76 |
"steps": 130,
|
77 |
"time": "20:36",
|
78 |
+
"rank": 4
|
79 |
},
|
80 |
{
|
81 |
"model": "deepseek-v3",
|
82 |
"score": 256,
|
83 |
"steps": 216,
|
84 |
"time": "54.02",
|
85 |
+
"rank": 5
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"model": "gemini-2.5-flash-preview-04-17",
|
89 |
+
"score": 128,
|
90 |
+
"steps": 71,
|
91 |
+
"time": "41:42",
|
92 |
+
"rank": 6
|
93 |
},
|
94 |
{
|
95 |
"model": "gemini-2.0-flash",
|
96 |
"score": 128,
|
97 |
"steps": 111,
|
98 |
"time": "18:43",
|
99 |
+
"rank": 7
|
100 |
},
|
101 |
{
|
102 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
103 |
"score": 128,
|
104 |
"steps": 132,
|
105 |
"time": ">100",
|
106 |
+
"rank": 8
|
107 |
},
|
108 |
{
|
109 |
"model": "gemini-2.5-pro-exp-03-25",
|
110 |
"score": 128,
|
111 |
"steps": 138,
|
112 |
"time": "169",
|
113 |
+
"rank": 9
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
117 |
+
"score": 128,
|
118 |
+
"steps": 145,
|
119 |
+
"time": ">100",
|
120 |
+
"rank": 10
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"model": "o4-mini",
|
124 |
+
"score": 128,
|
125 |
+
"steps": "",
|
126 |
+
"time": "",
|
127 |
+
"rank": 11
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"model": "grok3-beta",
|
131 |
+
"score": 128,
|
132 |
+
"steps": "",
|
133 |
+
"time": "",
|
134 |
+
"rank": 12
|
135 |
},
|
136 |
{
|
137 |
"model": "claude-3-5-sonnet-20241022",
|
138 |
"score": 64,
|
139 |
"steps": 92,
|
140 |
"time": "9:2",
|
141 |
+
"rank": 13
|
142 |
},
|
143 |
{
|
144 |
"model": "gpt-4.5-preview-2025-02-27",
|
145 |
"score": 34,
|
146 |
"steps": 34,
|
147 |
"time": "8:25",
|
148 |
+
"rank": 14
|
149 |
},
|
150 |
{
|
151 |
"model": "gpt-4o-2024-11-20",
|
152 |
"score": 16,
|
153 |
"steps": 21,
|
154 |
"time": "1:17",
|
155 |
+
"rank": 15
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
156 |
}
|
157 |
]
|
158 |
},
|
|
|
217 |
"Candy Crush": {
|
218 |
"runs": 3,
|
219 |
"results": [
|
220 |
+
{
|
221 |
+
"model": "o4-mini",
|
222 |
+
"score_runs": "123,131",
|
223 |
+
"average_score": 127,
|
224 |
+
"steps": 25,
|
225 |
+
"rank": 1
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"model": "o3",
|
229 |
+
"score_runs": "115, 122",
|
230 |
+
"average_score": 118.5,
|
231 |
+
"steps": 25,
|
232 |
+
"rank": 2
|
233 |
+
},
|
234 |
{
|
235 |
"model": "o3-mini-2025-01-31(medium)",
|
236 |
"score_runs": "90;109;120",
|
237 |
"average_score": 106.33,
|
238 |
"steps": 25,
|
239 |
+
"rank": 3
|
240 |
},
|
241 |
{
|
242 |
"model": "o1-2024-12-17",
|
243 |
"score_runs": "96;114;83",
|
244 |
"average_score": 97.67,
|
245 |
"steps": 25,
|
246 |
+
"rank": 4
|
247 |
},
|
248 |
{
|
249 |
"model": "deepseek-r1",
|
250 |
"score_runs": "62;108;105",
|
251 |
"average_score": 91.67,
|
252 |
"steps": 25,
|
253 |
+
"rank": 5
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"model": "gemini-2.5-flash-preview-04-17",
|
257 |
+
"score_runs": "59",
|
258 |
+
"average_score": 59,
|
259 |
+
"steps": 25,
|
260 |
+
"rank": 6
|
261 |
},
|
262 |
{
|
263 |
"model": "gemini-2.5-pro-exp-03-25",
|
264 |
"score_runs": "50;36;68",
|
265 |
"average_score": 51.33,
|
266 |
"steps": 25,
|
267 |
+
"rank": 7
|
268 |
},
|
269 |
{
|
270 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
271 |
"score_runs": "36;46;24",
|
272 |
"average_score": 35.33,
|
273 |
"steps": 25,
|
274 |
+
"rank": 8
|
275 |
},
|
276 |
{
|
277 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
278 |
"score_runs": "0;15;39",
|
279 |
"average_score": 18,
|
280 |
"steps": 25,
|
281 |
+
"rank": 9
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"model": "grok-3-beta",
|
285 |
+
"score_runs": "11",
|
286 |
+
"average_score": 11,
|
287 |
+
"steps": 25,
|
288 |
+
"rank": 10
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
292 |
+
"score_runs": "6;0;0",
|
293 |
+
"average_score": 2,
|
294 |
+
"steps": 25,
|
295 |
+
"rank": 11
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"model": "gpt-4.1-2025-04-14",
|
299 |
+
"score_runs": "0;3;3",
|
300 |
+
"average_score": 2,
|
301 |
+
"steps": 25,
|
302 |
+
"rank": 12
|
303 |
},
|
304 |
{
|
305 |
"model": "claude-3-5-sonnet-20241022",
|
306 |
"score_runs": "3;0;0",
|
307 |
"average_score": 1,
|
308 |
"steps": 25,
|
309 |
+
"rank": 13
|
310 |
},
|
311 |
{
|
312 |
"model": "deepseek-v3",
|
313 |
"score_runs": "0;0;0",
|
314 |
"average_score": 0,
|
315 |
"steps": 25,
|
316 |
+
"rank": 14
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
317 |
}
|
318 |
]
|
319 |
},
|
320 |
"Sokoban": {
|
321 |
"runs": 3,
|
322 |
"results": [
|
323 |
+
{
|
324 |
+
"model": "o3",
|
325 |
+
"levels_cracked": "4",
|
326 |
+
"steps": "[16, 40, 59, 110]",
|
327 |
+
"rank": 1
|
328 |
+
},
|
329 |
{
|
330 |
"model": "o3-mini-2025-01-31(medium)",
|
331 |
"levels_cracked": "2; 3; 2",
|
332 |
"steps": "[17,52,68];[24,58,78,91];[19,44,64]",
|
333 |
+
"rank": 2
|
334 |
},
|
335 |
{
|
336 |
"model": "gemini-2.5-pro-exp-03-25",
|
337 |
"levels_cracked": "2;2;3",
|
338 |
"steps": "[23, 46, 79]; [20,50,77]; [26,95,125,175]",
|
339 |
+
"rank": 3
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"model": "gemini-2.5-flash-preview-04-17",
|
343 |
+
"levels_cracked": "2",
|
344 |
+
"steps": "[24, 50, 60]",
|
345 |
+
"rank": 4
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"model": "o4-mini",
|
349 |
+
"levels_cracked": "2",
|
350 |
+
"steps": "",
|
351 |
+
"rank": 5
|
352 |
},
|
353 |
{
|
354 |
"model": "claude-3-7-sonnet-20250219(thinking)",
|
355 |
"levels_cracked": "1; 2; 0",
|
356 |
"steps": "[17,35];[15,40,43];[4]",
|
357 |
+
"rank": 6
|
358 |
},
|
359 |
{
|
360 |
"model": "o1-2024-12-17",
|
361 |
"levels_cracked": "1; 1; 1",
|
362 |
"steps": null,
|
363 |
+
"rank": 7
|
364 |
},
|
365 |
{
|
366 |
"model": "deepseek-r1",
|
367 |
"levels_cracked": "1; 0; 1",
|
368 |
"steps": "[19,42];[13];[19,36]",
|
369 |
"note": "stuck",
|
370 |
+
"rank": 8
|
371 |
},
|
372 |
{
|
373 |
"model": "o1-mini-2024-09-12",
|
374 |
"levels_cracked": "0;1;0",
|
375 |
"steps": null,
|
376 |
+
"rank": 9
|
377 |
},
|
378 |
{
|
379 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
380 |
"levels_cracked": "0; 0; 0",
|
381 |
"steps": "[23]; [14]; [14]",
|
382 |
+
"rank": 10
|
383 |
},
|
384 |
{
|
385 |
"model": "gpt-4o-2024-11-20",
|
386 |
"levels_cracked": "0; 0; 0",
|
387 |
"steps": "[68];[105];[168]",
|
388 |
"note": "stuck in a loop",
|
389 |
+
"rank": 11
|
390 |
},
|
391 |
{
|
392 |
"model": "claude-3-5-sonnet-20241022",
|
393 |
"levels_cracked": "0; 0; 0",
|
394 |
"steps": "[21]; [30]; [51]",
|
395 |
"note": "stuck in a loop",
|
396 |
+
"rank": 12
|
397 |
},
|
398 |
{
|
399 |
"model": "deepseek-v3",
|
400 |
"levels_cracked": "0; 0; 0",
|
401 |
"steps": "[9]; [47]; [64]",
|
402 |
+
"rank": 13
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"model": "gpt-4.1-2025-04-14",
|
406 |
+
"levels_cracked": "0; 0; 0",
|
407 |
+
"steps": "[9]; [47]; [64]",
|
408 |
+
"rank": 14
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"model": "grok-3-beta",
|
412 |
+
"levels_cracked": "0",
|
413 |
+
"steps": "",
|
414 |
+
"rank": 15
|
415 |
},
|
416 |
{
|
417 |
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
418 |
"levels_cracked": "0;0;0",
|
419 |
"steps": "[5]",
|
420 |
+
"rank": 16
|
421 |
}
|
422 |
]
|
423 |
},
|
|
|
433 |
"score": 26,
|
434 |
"note": "stuck at the end not present evidence"
|
435 |
},
|
436 |
+
{
|
437 |
+
"model": "o3",
|
438 |
+
"levels_cracked": "3",
|
439 |
+
"lives_left": "[5, 3, 3, 0]",
|
440 |
+
"cracked_details": "4: 4/8",
|
441 |
+
"rank": 2,
|
442 |
+
"score": 23,
|
443 |
+
"note": "failed to present evidence"
|
444 |
+
},
|
445 |
{
|
446 |
"model": "gemini-2.5-pro-exp-03-25",
|
447 |
"levels_cracked": "2; 3",
|
448 |
"lives_left": "[5,5,0]; [5, 5, 4, 0]",
|
449 |
"cracked_details": "4: 0/8",
|
450 |
+
"rank": 3,
|
451 |
"score": 20,
|
452 |
"note": "failed to present evidence"
|
453 |
},
|
|
|
456 |
"levels_cracked": "1; 1",
|
457 |
"lives_left": "[3,0]; [5,0]",
|
458 |
"cracked_details": "2: 3/9",
|
459 |
+
"rank": 4,
|
460 |
"score": 8,
|
461 |
"note": "failed to present evidence"
|
462 |
},
|
|
|
465 |
"levels_cracked": "1",
|
466 |
"lives_left": "5, 5",
|
467 |
"cracked_details": "1:1/8",
|
468 |
+
"rank": 5,
|
469 |
"score": 6,
|
470 |
"note": "stuck in loop"
|
471 |
},
|
|
|
474 |
"levels_cracked": "1",
|
475 |
"lives_left": "[4,5]",
|
476 |
"cracked_details": "1: 1/8",
|
477 |
+
"rank": 6,
|
478 |
"score": 6,
|
479 |
"note": "stuck in loop"
|
480 |
},
|
481 |
+
{
|
482 |
+
"model": "gemini-2.5-flash-preview-04-17",
|
483 |
+
"levels_cracked": "0",
|
484 |
+
"lives_left": "0",
|
485 |
+
"cracked_details": "1: 4/5",
|
486 |
+
"rank": 7,
|
487 |
+
"score": 4,
|
488 |
+
"note": "stuck in the last option section"
|
489 |
+
},
|
490 |
{
|
491 |
"model": "gemini-2.0-flash-thinking-exp-1219",
|
492 |
"levels_cracked": "0",
|
493 |
"lives_left": "0",
|
494 |
"cracked_details": "1: 4/5",
|
495 |
+
"rank": 8,
|
496 |
"score": 4,
|
497 |
"note": "stuck in the last option section"
|
498 |
},
|
|
|
501 |
"levels_cracked": "0",
|
502 |
"lives_left": "0",
|
503 |
"cracked_details": "1: 4/5",
|
504 |
+
"rank": 9,
|
505 |
"score": 4,
|
506 |
"note": "stuck in the 3rd evidence present"
|
507 |
},
|
508 |
+
{
|
509 |
+
"model": "o4-mini",
|
510 |
+
"levels_cracked": "0",
|
511 |
+
"lives_left": "0",
|
512 |
+
"cracked_details": "1:1/5",
|
513 |
+
"rank": 10,
|
514 |
+
"score": 1,
|
515 |
+
"note": "failed to present evidence"
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"model": "grok-3-beta",
|
519 |
+
"levels_cracked": "0",
|
520 |
+
"lives_left": "0",
|
521 |
+
"cracked_details": "1:1/5",
|
522 |
+
"rank": 11,
|
523 |
+
"score": 1,
|
524 |
+
"note": "failed to present evidence"
|
525 |
+
},
|
526 |
{
|
527 |
"model": "Llama-4-Maverick-17B-128E-Instruct-FP8",
|
528 |
"levels_cracked": "0",
|
529 |
"lives_left": "0",
|
530 |
"cracked_details": "0:0/5",
|
531 |
+
"rank": 12,
|
532 |
"score": 0,
|
533 |
"note": "failed to present evidence"
|
534 |
}
|