CodeGoat24 commited on
Commit
92a7023
·
verified ·
1 Parent(s): f4ec9f4

Update leaderboard_data.json

Browse files
Files changed (1) hide show
  1. leaderboard_data.json +450 -450
leaderboard_data.json CHANGED
@@ -1,456 +1,456 @@
1
  {
2
  "leaderboard": [
3
  {
4
- "model": "FLUX-2-flex",
5
- "link": "https://flux2-flex.com/",
6
- "hf": "-",
7
- "open_source": false,
8
- "release_date": "2025-11",
9
- "Overall": 90.43,
10
- "Style": 98.73,
11
- "World Knowledge": 97.02,
12
- "Attribute-Overall": 94.55,
13
- "Quantity": 90.43,
14
- "Expression": 88.65,
15
- "Material": 95.15,
16
- "Size": 93.65,
17
- "Shape": 91.33,
18
- "Color": 97.93,
19
- "Action-Overall": 85.74,
20
- "Hand": 85.67,
21
- "Full body": 86.86,
22
- "Animal": 84.78,
23
- "Non Contact": 83.71,
24
- "Contact": 82.06,
25
- "State": 87.36,
26
- "Relationship-Overall": 90.33,
27
- "Composition": 88.72,
28
- "Similarity": 88.2,
29
- "Inclusion": 93.6,
30
- "Comparison": 92.23,
31
- "Compound-Overall": 91.2,
32
- "Imagination": 91.95,
33
- "Feature matching": 89.52,
34
- "Grammar-Overall": 86.78,
35
- "Pronoun Reference": 96.61,
36
- "Consistency": 82.66,
37
- "Negation": 82.14,
38
- "Layout-Overall": 93.82,
39
- "2D": 93.86,
40
- "3D": 93.76,
41
- "Logical Reasoning": 75.74,
42
- "Text": 90.38
43
- },
44
- {
45
- "model": "FLUX-2-max",
46
- "link": "https://bfl.ai/models/flux-2-max",
47
- "hf": "-",
48
- "open_source": false,
49
- "release_date": "2025-12",
50
- "Overall": 92.18,
51
- "Style": 99.24,
52
- "World Knowledge": 96.73,
53
- "Attribute-Overall": 94.78,
54
- "Quantity": 88.3,
55
- "Expression": 86.06,
56
- "Material": 96.36,
57
- "Size": 94.06,
58
- "Shape": 93.11,
59
- "Color": 98.25,
60
- "Action-Overall": 86.97,
61
- "Hand": 84.67,
62
- "Full body": 88.78,
63
- "Animal": 88.77,
64
- "Non Contact": 84.23,
65
- "Contact": 84.12,
66
- "State": 88.17,
67
- "Relationship-Overall": 92.9,
68
- "Composition": 92.31,
69
- "Similarity": 90.58,
70
- "Inclusion": 96.51,
71
- "Comparison": 92.72,
72
- "Compound-Overall": 92.84,
73
- "Imagination": 93.04,
74
- "Feature matching": 92.38,
75
- "Grammar-Overall": 89.4,
76
- "Pronoun Reference": 96.61,
77
- "Consistency": 88.31,
78
- "Negation": 84.29,
79
- "Layout-Overall": 95.02,
80
- "2D": 95.57,
81
- "3D": 94.32,
82
- "Logical Reasoning": 83.0,
83
- "Text": 90.93
84
- },
85
- {
86
- "model": "FLUX-2-pro",
87
- "link": "https://docs.bfl.ai/flux_2/flux2_text_to_image",
88
- "hf": "-",
89
- "open_source": false,
90
- "release_date": "2025-11",
91
- "Overall": 90.1,
92
- "Style": 99.08,
93
- "World Knowledge": 96.89,
94
- "Attribute-Overall": 94.37,
95
- "Quantity": 86.7,
96
- "Expression": 86.93,
97
- "Material": 96.67,
98
- "Size": 92.94,
99
- "Shape": 91.86,
100
- "Color": 97.41,
101
- "Action-Overall": 84.38,
102
- "Hand": 81.0,
103
- "Full body": 85.76,
104
- "Animal": 83.7,
105
- "Non Contact": 82.35,
106
- "Contact": 80.0,
107
- "State": 87.06,
108
- "Relationship-Overall": 90.86,
109
- "Composition": 89.92,
110
- "Similarity": 87.7,
111
- "Inclusion": 95.35,
112
- "Comparison": 91.26,
113
- "Compound-Overall": 90.81,
114
- "Imagination": 91.67,
115
- "Feature matching": 88.92,
116
- "Grammar-Overall": 87.83,
117
- "Pronoun Reference": 94.49,
118
- "Consistency": 87.9,
119
- "Negation": 82.14,
120
- "Layout-Overall": 93.98,
121
- "2D": 96.14,
122
- "3D": 91.19,
123
- "Logical Reasoning": 80.15,
124
- "Text": 82.69
125
- },
126
- {
127
- "model": "FLUX.2-klein-4b",
128
- "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
129
- "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-4B",
130
- "open_source": true,
131
- "release_date": "2026-1",
132
- "Overall": 80.4,
133
- "Style": 98.01,
134
- "World Knowledge": 91.18,
135
- "Attribute-Overall": 93.06,
136
- "Quantity": 75.53,
137
- "Expression": 89.03,
138
- "Material": 94.95,
139
- "Size": 88.45,
140
- "Shape": 87.06,
141
- "Color": 98.06,
142
- "Action-Overall": 79.78,
143
- "Hand": 75.96,
144
- "Full body": 77.19,
145
- "Animal": 81.52,
146
- "Non Contact": 77.23,
147
- "Contact": 72.41,
148
- "State": 84.42,
149
- "Relationship-Overall": 87.45,
150
- "Composition": 88.01,
151
- "Similarity": 80.13,
152
- "Inclusion": 93.39,
153
- "Comparison": 86.89,
154
- "Compound-Overall": 85.2,
155
- "Imagination": 84.64,
156
- "Feature matching": 86.45,
157
- "Grammar-Overall": 85.41,
158
- "Pronoun Reference": 93.25,
159
- "Consistency": 82.54,
160
- "Negation": 80.99,
161
- "Layout-Overall": 89.52,
162
- "2D": 90.96,
163
- "3D": 87.68,
164
- "Logical Reasoning": 74.02,
165
- "Text": 20.38
166
- },
167
- {
168
- "model": "FLUX.2-klein-9b",
169
- "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
170
- "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-9B",
171
- "open_source": true,
172
- "release_date": "2026-1",
173
- "Overall": 85.06,
174
- "Style": 98.67,
175
- "World Knowledge": 94.65,
176
- "Attribute-Overall": 94.11,
177
- "Quantity": 80.85,
178
- "Expression": 88.06,
179
- "Material": 95.56,
180
- "Size": 91.48,
181
- "Shape": 89.69,
182
- "Color": 98.74,
183
- "Action-Overall": 82.4,
184
- "Hand": 80.13,
185
- "Full body": 81.56,
186
- "Animal": 83.7,
187
- "Non Contact": 78.57,
188
- "Contact": 76.44,
189
- "State": 85.91,
190
- "Relationship-Overall": 89.92,
191
- "Composition": 90.69,
192
- "Similarity": 83.33,
193
- "Inclusion": 94.83,
194
- "Comparison": 89.32,
195
- "Compound-Overall": 88.7,
196
- "Imagination": 89.51,
197
- "Feature matching": 86.92,
198
- "Grammar-Overall": 86.68,
199
- "Pronoun Reference": 93.65,
200
- "Consistency": 86.11,
201
- "Negation": 80.99,
202
- "Layout-Overall": 93.17,
203
- "2D": 93.22,
204
- "3D": 93.12,
205
- "Logical Reasoning": 75.25,
206
- "Text": 47.01
207
- },
208
- {
209
- "model": "FLUX.2-klein-base-4b",
210
- "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
211
- "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B",
212
- "open_source": true,
213
- "release_date": "2026-1",
214
- "Overall": 77.76,
215
- "Style": 97.67,
216
- "World Knowledge": 87.72,
217
- "Attribute-Overall": 90.54,
218
- "Quantity": 76.06,
219
- "Expression": 81.39,
220
- "Material": 91.79,
221
- "Size": 85.8,
222
- "Shape": 86.19,
223
- "Color": 97.17,
224
- "Action-Overall": 75.24,
225
- "Hand": 74.04,
226
- "Full body": 71.25,
227
- "Animal": 76.09,
228
- "Non Contact": 72.77,
229
- "Contact": 65.8,
230
- "State": 80.46,
231
- "Relationship-Overall": 83.14,
232
- "Composition": 82.91,
233
- "Similarity": 78.53,
234
- "Inclusion": 85.92,
235
- "Comparison": 84.71,
236
- "Compound-Overall": 81.92,
237
- "Imagination": 80.61,
238
- "Feature matching": 84.81,
239
- "Grammar-Overall": 86.55,
240
- "Pronoun Reference": 92.46,
241
- "Consistency": 84.13,
242
- "Negation": 83.45,
243
- "Layout-Overall": 89.13,
244
- "2D": 92.23,
245
- "3D": 85.14,
246
- "Logical Reasoning": 68.87,
247
- "Text": 16.85
248
- },
249
- {
250
- "model": "FLUX.2-klein-base-9b",
251
- "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
252
- "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B",
253
- "open_source": true,
254
- "release_date": "2026-1",
255
- "Overall": 86.45,
256
- "Style": 97.92,
257
- "World Knowledge": 95.38,
258
- "Attribute-Overall": 92.79,
259
- "Quantity": 79.79,
260
- "Expression": 86.67,
261
- "Material": 94.2,
262
- "Size": 90.53,
263
- "Shape": 87.24,
264
- "Color": 97.69,
265
- "Action-Overall": 80.83,
266
- "Hand": 81.73,
267
- "Full body": 80.94,
268
- "Animal": 80.43,
269
- "Non Contact": 76.34,
270
- "Contact": 73.56,
271
- "State": 84.13,
272
- "Relationship-Overall": 88.85,
273
- "Composition": 88.14,
274
- "Similarity": 87.5,
275
- "Inclusion": 95.69,
276
- "Comparison": 85.44,
277
- "Compound-Overall": 88.48,
278
- "Imagination": 88.98,
279
- "Feature matching": 87.38,
280
- "Grammar-Overall": 87.82,
281
- "Pronoun Reference": 93.65,
282
- "Consistency": 85.71,
283
- "Negation": 84.51,
284
- "Layout-Overall": 92.78,
285
- "2D": 92.66,
286
- "3D": 92.93,
287
- "Logical Reasoning": 77.94,
288
- "Text": 61.68
289
- },
290
- {
291
- "model": "GPT-4o-1.5",
292
- "link": "https://developers.openai.com/api/docs/models/gpt-image-1.5",
293
- "hf": "-",
294
- "open_source": false,
295
- "release_date": "2025-12",
296
- "Overall": 95.41,
297
- "Style": 99.58,
298
- "World Knowledge": 98.98,
299
- "Attribute-Overall": 97.2,
300
- "Quantity": 93.41,
301
- "Expression": 95.19,
302
- "Material": 97.25,
303
- "Size": 95.39,
304
- "Shape": 95.98,
305
- "Color": 99.15,
306
- "Action-Overall": 92.9,
307
- "Hand": 92.33,
308
- "Full body": 93.99,
309
- "Animal": 95.29,
310
- "Non Contact": 91.71,
311
- "Contact": 92.51,
312
- "State": 92.46,
313
- "Relationship-Overall": 95.79,
314
- "Composition": 96.11,
315
- "Similarity": 94.16,
316
- "Inclusion": 99.71,
317
- "Comparison": 93.15,
318
- "Compound-Overall": 96.45,
319
- "Imagination": 97.54,
320
- "Feature matching": 93.98,
321
- "Grammar-Overall": 94.84,
322
- "Pronoun Reference": 95.16,
323
- "Consistency": 95.24,
324
- "Negation": 94.18,
325
- "Layout-Overall": 96.7,
326
- "2D": 96.71,
327
- "3D": 96.69,
328
- "Logical Reasoning": 90.15,
329
- "Text": 91.46
330
- },
331
- {
332
- "model": "LongCat-Image",
333
- "link": "https://arxiv.org/pdf/2512.07584",
334
- "hf": "https://huggingface.co/meituan-longcat/LongCat-Image",
335
- "open_source": true,
336
- "release_date": "2025-12",
337
- "Overall": 81.28,
338
- "Style": 92.11,
339
- "World Knowledge": 93.5,
340
- "Attribute-Overall": 90.01,
341
- "Quantity": 77.13,
342
- "Expression": 82.22,
343
- "Material": 91.79,
344
- "Size": 88.64,
345
- "Shape": 80.94,
346
- "Color": 96.07,
347
- "Action-Overall": 77.69,
348
- "Hand": 73.72,
349
- "Full body": 73.44,
350
- "Animal": 79.35,
351
- "Non Contact": 74.11,
352
- "Contact": 66.95,
353
- "State": 84.33,
354
- "Relationship-Overall": 81.3,
355
- "Composition": 79.85,
356
- "Similarity": 75.0,
357
- "Inclusion": 89.08,
358
- "Comparison": 82.28,
359
- "Compound-Overall": 79.15,
360
- "Imagination": 79.98,
361
- "Feature matching": 77.34,
362
- "Grammar-Overall": 75.89,
363
- "Pronoun Reference": 91.27,
364
- "Consistency": 71.83,
365
- "Negation": 65.85,
366
- "Layout-Overall": 87.22,
367
- "2D": 89.55,
368
- "3D": 84.24,
369
- "Logical Reasoning": 66.91,
370
- "Text": 69.02
371
- },
372
- {
373
- "model": "Seedream-4-5-251128",
374
- "link": "https://seed.bytedance.com/en/seedream4_5",
375
- "hf": "-",
376
- "open_source": false,
377
- "release_date": "2025-11",
378
- "Overall": 91.38,
379
- "Style": 98.67,
380
- "World Knowledge": 96.24,
381
- "Attribute-Overall": 96.15,
382
- "Quantity": 90.43,
383
- "Expression": 91.38,
384
- "Material": 95.56,
385
- "Size": 94.89,
386
- "Shape": 96.5,
387
- "Color": 99.16,
388
- "Action-Overall": 88.2,
389
- "Hand": 83.97,
390
- "Full body": 90.31,
391
- "Animal": 88.73,
392
- "Non Contact": 87.89,
393
- "Contact": 83.0,
394
- "State": 90.57,
395
- "Relationship-Overall": 89.92,
396
- "Composition": 88.39,
397
- "Similarity": 84.89,
398
- "Inclusion": 95.98,
399
- "Comparison": 91.5,
400
- "Compound-Overall": 89.57,
401
- "Imagination": 89.71,
402
- "Feature matching": 89.25,
403
- "Grammar-Overall": 86.8,
404
- "Pronoun Reference": 97.62,
405
- "Consistency": 88.89,
406
- "Negation": 75.35,
407
- "Layout-Overall": 93.33,
408
- "2D": 93.36,
409
- "3D": 93.28,
410
- "Logical Reasoning": 83.09,
411
- "Text": 91.85
412
- },
413
- {
414
- "model": "Z-Image",
415
- "link": "https://arxiv.org/pdf/2511.22699",
416
- "hf": "https://huggingface.co/Tongyi-MAI/Z-Image",
417
- "open_source": true,
418
- "release_date": "2026-1",
419
- "Overall": 86.77,
420
- "Style": 97.26,
421
- "World Knowledge": 94.36,
422
- "Attribute-Overall": 93.25,
423
- "Quantity": 85.11,
424
- "Expression": 87.08,
425
- "Material": 94.95,
426
- "Size": 90.15,
427
- "Shape": 87.41,
428
- "Color": 97.8,
429
- "Action-Overall": 83.72,
430
- "Hand": 81.73,
431
- "Full body": 79.38,
432
- "Animal": 83.33,
433
- "Non Contact": 86.16,
434
- "Contact": 77.87,
435
- "State": 87.3,
436
- "Relationship-Overall": 89.06,
437
- "Composition": 88.52,
438
- "Similarity": 84.94,
439
- "Inclusion": 93.39,
440
- "Comparison": 89.56,
441
- "Compound-Overall": 86.52,
442
- "Imagination": 86.12,
443
- "Feature matching": 87.38,
444
- "Grammar-Overall": 80.46,
445
- "Pronoun Reference": 91.67,
446
- "Consistency": 80.16,
447
- "Negation": 70.77,
448
- "Layout-Overall": 91.11,
449
- "2D": 92.51,
450
- "3D": 89.31,
451
- "Logical Reasoning": 76.72,
452
- "Text": 75.27
453
- },
454
  {
455
  "model": "FLUX.2-dev",
456
  "link": "https://github.com/black-forest-labs/flux2",
 
1
  {
2
  "leaderboard": [
3
  {
4
+ "model": "FLUX-2-flex",
5
+ "link": "https://flux2-flex.com/",
6
+ "hf": "-",
7
+ "open_source": false,
8
+ "release_date": "2025-11",
9
+ "Overall": 90.43,
10
+ "Style": 98.73,
11
+ "World Knowledge": 97.02,
12
+ "Attribute-Overall": 94.55,
13
+ "Quantity": 90.43,
14
+ "Expression": 88.65,
15
+ "Material": 95.15,
16
+ "Size": 93.65,
17
+ "Shape": 91.33,
18
+ "Color": 97.93,
19
+ "Action-Overall": 85.74,
20
+ "Hand": 85.67,
21
+ "Full body": 86.86,
22
+ "Animal": 84.78,
23
+ "Non Contact": 83.71,
24
+ "Contact": 82.06,
25
+ "State": 87.36,
26
+ "Relationship-Overall": 90.33,
27
+ "Composition": 88.72,
28
+ "Similarity": 88.2,
29
+ "Inclusion": 93.6,
30
+ "Comparison": 92.23,
31
+ "Compound-Overall": 91.2,
32
+ "Imagination": 91.95,
33
+ "Feature matching": 89.52,
34
+ "Grammar-Overall": 86.78,
35
+ "Pronoun Reference": 96.61,
36
+ "Consistency": 82.66,
37
+ "Negation": 82.14,
38
+ "Layout-Overall": 93.82,
39
+ "2D": 93.86,
40
+ "3D": 93.76,
41
+ "Logical Reasoning": 75.74,
42
+ "Text": 90.38
43
+ },
44
+ {
45
+ "model": "FLUX-2-max",
46
+ "link": "https://bfl.ai/models/flux-2-max",
47
+ "hf": "-",
48
+ "open_source": false,
49
+ "release_date": "2025-12",
50
+ "Overall": 92.18,
51
+ "Style": 99.24,
52
+ "World Knowledge": 96.73,
53
+ "Attribute-Overall": 94.78,
54
+ "Quantity": 88.3,
55
+ "Expression": 86.06,
56
+ "Material": 96.36,
57
+ "Size": 94.06,
58
+ "Shape": 93.11,
59
+ "Color": 98.25,
60
+ "Action-Overall": 86.97,
61
+ "Hand": 84.67,
62
+ "Full body": 88.78,
63
+ "Animal": 88.77,
64
+ "Non Contact": 84.23,
65
+ "Contact": 84.12,
66
+ "State": 88.17,
67
+ "Relationship-Overall": 92.9,
68
+ "Composition": 92.31,
69
+ "Similarity": 90.58,
70
+ "Inclusion": 96.51,
71
+ "Comparison": 92.72,
72
+ "Compound-Overall": 92.84,
73
+ "Imagination": 93.04,
74
+ "Feature matching": 92.38,
75
+ "Grammar-Overall": 89.4,
76
+ "Pronoun Reference": 96.61,
77
+ "Consistency": 88.31,
78
+ "Negation": 84.29,
79
+ "Layout-Overall": 95.02,
80
+ "2D": 95.57,
81
+ "3D": 94.32,
82
+ "Logical Reasoning": 83.0,
83
+ "Text": 90.93
84
+ },
85
+ {
86
+ "model": "FLUX-2-pro",
87
+ "link": "https://docs.bfl.ai/flux_2/flux2_text_to_image",
88
+ "hf": "-",
89
+ "open_source": false,
90
+ "release_date": "2025-11",
91
+ "Overall": 90.1,
92
+ "Style": 99.08,
93
+ "World Knowledge": 96.89,
94
+ "Attribute-Overall": 94.37,
95
+ "Quantity": 86.7,
96
+ "Expression": 86.93,
97
+ "Material": 96.67,
98
+ "Size": 92.94,
99
+ "Shape": 91.86,
100
+ "Color": 97.41,
101
+ "Action-Overall": 84.38,
102
+ "Hand": 81.0,
103
+ "Full body": 85.76,
104
+ "Animal": 83.7,
105
+ "Non Contact": 82.35,
106
+ "Contact": 80.0,
107
+ "State": 87.06,
108
+ "Relationship-Overall": 90.86,
109
+ "Composition": 89.92,
110
+ "Similarity": 87.7,
111
+ "Inclusion": 95.35,
112
+ "Comparison": 91.26,
113
+ "Compound-Overall": 90.81,
114
+ "Imagination": 91.67,
115
+ "Feature matching": 88.92,
116
+ "Grammar-Overall": 87.83,
117
+ "Pronoun Reference": 94.49,
118
+ "Consistency": 87.9,
119
+ "Negation": 82.14,
120
+ "Layout-Overall": 93.98,
121
+ "2D": 96.14,
122
+ "3D": 91.19,
123
+ "Logical Reasoning": 80.15,
124
+ "Text": 82.69
125
+ },
126
+ {
127
+ "model": "FLUX.2-klein-4b",
128
+ "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
129
+ "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-4B",
130
+ "open_source": true,
131
+ "release_date": "2026-1",
132
+ "Overall": 80.4,
133
+ "Style": 98.01,
134
+ "World Knowledge": 91.18,
135
+ "Attribute-Overall": 93.06,
136
+ "Quantity": 75.53,
137
+ "Expression": 89.03,
138
+ "Material": 94.95,
139
+ "Size": 88.45,
140
+ "Shape": 87.06,
141
+ "Color": 98.06,
142
+ "Action-Overall": 79.78,
143
+ "Hand": 75.96,
144
+ "Full body": 77.19,
145
+ "Animal": 81.52,
146
+ "Non Contact": 77.23,
147
+ "Contact": 72.41,
148
+ "State": 84.42,
149
+ "Relationship-Overall": 87.45,
150
+ "Composition": 88.01,
151
+ "Similarity": 80.13,
152
+ "Inclusion": 93.39,
153
+ "Comparison": 86.89,
154
+ "Compound-Overall": 85.2,
155
+ "Imagination": 84.64,
156
+ "Feature matching": 86.45,
157
+ "Grammar-Overall": 85.41,
158
+ "Pronoun Reference": 93.25,
159
+ "Consistency": 82.54,
160
+ "Negation": 80.99,
161
+ "Layout-Overall": 89.52,
162
+ "2D": 90.96,
163
+ "3D": 87.68,
164
+ "Logical Reasoning": 74.02,
165
+ "Text": 20.38
166
+ },
167
+ {
168
+ "model": "FLUX.2-klein-9b",
169
+ "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
170
+ "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-9B",
171
+ "open_source": true,
172
+ "release_date": "2026-1",
173
+ "Overall": 85.06,
174
+ "Style": 98.67,
175
+ "World Knowledge": 94.65,
176
+ "Attribute-Overall": 94.11,
177
+ "Quantity": 80.85,
178
+ "Expression": 88.06,
179
+ "Material": 95.56,
180
+ "Size": 91.48,
181
+ "Shape": 89.69,
182
+ "Color": 98.74,
183
+ "Action-Overall": 82.4,
184
+ "Hand": 80.13,
185
+ "Full body": 81.56,
186
+ "Animal": 83.7,
187
+ "Non Contact": 78.57,
188
+ "Contact": 76.44,
189
+ "State": 85.91,
190
+ "Relationship-Overall": 89.92,
191
+ "Composition": 90.69,
192
+ "Similarity": 83.33,
193
+ "Inclusion": 94.83,
194
+ "Comparison": 89.32,
195
+ "Compound-Overall": 88.7,
196
+ "Imagination": 89.51,
197
+ "Feature matching": 86.92,
198
+ "Grammar-Overall": 86.68,
199
+ "Pronoun Reference": 93.65,
200
+ "Consistency": 86.11,
201
+ "Negation": 80.99,
202
+ "Layout-Overall": 93.17,
203
+ "2D": 93.22,
204
+ "3D": 93.12,
205
+ "Logical Reasoning": 75.25,
206
+ "Text": 47.01
207
+ },
208
+ {
209
+ "model": "FLUX.2-klein-base-4b",
210
+ "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
211
+ "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-base-4B",
212
+ "open_source": true,
213
+ "release_date": "2026-1",
214
+ "Overall": 77.76,
215
+ "Style": 97.67,
216
+ "World Knowledge": 87.72,
217
+ "Attribute-Overall": 90.54,
218
+ "Quantity": 76.06,
219
+ "Expression": 81.39,
220
+ "Material": 91.79,
221
+ "Size": 85.8,
222
+ "Shape": 86.19,
223
+ "Color": 97.17,
224
+ "Action-Overall": 75.24,
225
+ "Hand": 74.04,
226
+ "Full body": 71.25,
227
+ "Animal": 76.09,
228
+ "Non Contact": 72.77,
229
+ "Contact": 65.8,
230
+ "State": 80.46,
231
+ "Relationship-Overall": 83.14,
232
+ "Composition": 82.91,
233
+ "Similarity": 78.53,
234
+ "Inclusion": 85.92,
235
+ "Comparison": 84.71,
236
+ "Compound-Overall": 81.92,
237
+ "Imagination": 80.61,
238
+ "Feature matching": 84.81,
239
+ "Grammar-Overall": 86.55,
240
+ "Pronoun Reference": 92.46,
241
+ "Consistency": 84.13,
242
+ "Negation": 83.45,
243
+ "Layout-Overall": 89.13,
244
+ "2D": 92.23,
245
+ "3D": 85.14,
246
+ "Logical Reasoning": 68.87,
247
+ "Text": 16.85
248
+ },
249
+ {
250
+ "model": "FLUX.2-klein-base-9b",
251
+ "link": "https://bfl.ai/blog/flux2-klein-towards-interactive-visual-intelligence",
252
+ "hf": "https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B",
253
+ "open_source": true,
254
+ "release_date": "2026-1",
255
+ "Overall": 86.45,
256
+ "Style": 97.92,
257
+ "World Knowledge": 95.38,
258
+ "Attribute-Overall": 92.79,
259
+ "Quantity": 79.79,
260
+ "Expression": 86.67,
261
+ "Material": 94.2,
262
+ "Size": 90.53,
263
+ "Shape": 87.24,
264
+ "Color": 97.69,
265
+ "Action-Overall": 80.83,
266
+ "Hand": 81.73,
267
+ "Full body": 80.94,
268
+ "Animal": 80.43,
269
+ "Non Contact": 76.34,
270
+ "Contact": 73.56,
271
+ "State": 84.13,
272
+ "Relationship-Overall": 88.85,
273
+ "Composition": 88.14,
274
+ "Similarity": 87.5,
275
+ "Inclusion": 95.69,
276
+ "Comparison": 85.44,
277
+ "Compound-Overall": 88.48,
278
+ "Imagination": 88.98,
279
+ "Feature matching": 87.38,
280
+ "Grammar-Overall": 87.82,
281
+ "Pronoun Reference": 93.65,
282
+ "Consistency": 85.71,
283
+ "Negation": 84.51,
284
+ "Layout-Overall": 92.78,
285
+ "2D": 92.66,
286
+ "3D": 92.93,
287
+ "Logical Reasoning": 77.94,
288
+ "Text": 61.68
289
+ },
290
+ {
291
+ "model": "GPT-4o-1.5",
292
+ "link": "https://developers.openai.com/api/docs/models/gpt-image-1.5",
293
+ "hf": "-",
294
+ "open_source": false,
295
+ "release_date": "2025-12",
296
+ "Overall": 95.41,
297
+ "Style": 99.58,
298
+ "World Knowledge": 98.98,
299
+ "Attribute-Overall": 97.2,
300
+ "Quantity": 93.41,
301
+ "Expression": 95.19,
302
+ "Material": 97.25,
303
+ "Size": 95.39,
304
+ "Shape": 95.98,
305
+ "Color": 99.15,
306
+ "Action-Overall": 92.9,
307
+ "Hand": 92.33,
308
+ "Full body": 93.99,
309
+ "Animal": 95.29,
310
+ "Non Contact": 91.71,
311
+ "Contact": 92.51,
312
+ "State": 92.46,
313
+ "Relationship-Overall": 95.79,
314
+ "Composition": 96.11,
315
+ "Similarity": 94.16,
316
+ "Inclusion": 99.71,
317
+ "Comparison": 93.15,
318
+ "Compound-Overall": 96.45,
319
+ "Imagination": 97.54,
320
+ "Feature matching": 93.98,
321
+ "Grammar-Overall": 94.84,
322
+ "Pronoun Reference": 95.16,
323
+ "Consistency": 95.24,
324
+ "Negation": 94.18,
325
+ "Layout-Overall": 96.7,
326
+ "2D": 96.71,
327
+ "3D": 96.69,
328
+ "Logical Reasoning": 90.15,
329
+ "Text": 91.46
330
+ },
331
+ {
332
+ "model": "LongCat-Image",
333
+ "link": "https://arxiv.org/pdf/2512.07584",
334
+ "hf": "https://huggingface.co/meituan-longcat/LongCat-Image",
335
+ "open_source": true,
336
+ "release_date": "2025-12",
337
+ "Overall": 81.28,
338
+ "Style": 92.11,
339
+ "World Knowledge": 93.5,
340
+ "Attribute-Overall": 90.01,
341
+ "Quantity": 77.13,
342
+ "Expression": 82.22,
343
+ "Material": 91.79,
344
+ "Size": 88.64,
345
+ "Shape": 80.94,
346
+ "Color": 96.07,
347
+ "Action-Overall": 77.69,
348
+ "Hand": 73.72,
349
+ "Full body": 73.44,
350
+ "Animal": 79.35,
351
+ "Non Contact": 74.11,
352
+ "Contact": 66.95,
353
+ "State": 84.33,
354
+ "Relationship-Overall": 81.3,
355
+ "Composition": 79.85,
356
+ "Similarity": 75.0,
357
+ "Inclusion": 89.08,
358
+ "Comparison": 82.28,
359
+ "Compound-Overall": 79.15,
360
+ "Imagination": 79.98,
361
+ "Feature matching": 77.34,
362
+ "Grammar-Overall": 75.89,
363
+ "Pronoun Reference": 91.27,
364
+ "Consistency": 71.83,
365
+ "Negation": 65.85,
366
+ "Layout-Overall": 87.22,
367
+ "2D": 89.55,
368
+ "3D": 84.24,
369
+ "Logical Reasoning": 66.91,
370
+ "Text": 69.02
371
+ },
372
+ {
373
+ "model": "Seedream-4-5-251128",
374
+ "link": "https://seed.bytedance.com/en/seedream4_5",
375
+ "hf": "-",
376
+ "open_source": false,
377
+ "release_date": "2025-11",
378
+ "Overall": 91.38,
379
+ "Style": 98.67,
380
+ "World Knowledge": 96.24,
381
+ "Attribute-Overall": 96.15,
382
+ "Quantity": 90.43,
383
+ "Expression": 91.38,
384
+ "Material": 95.56,
385
+ "Size": 94.89,
386
+ "Shape": 96.5,
387
+ "Color": 99.16,
388
+ "Action-Overall": 88.2,
389
+ "Hand": 83.97,
390
+ "Full body": 90.31,
391
+ "Animal": 88.73,
392
+ "Non Contact": 87.89,
393
+ "Contact": 83.0,
394
+ "State": 90.57,
395
+ "Relationship-Overall": 89.92,
396
+ "Composition": 88.39,
397
+ "Similarity": 84.89,
398
+ "Inclusion": 95.98,
399
+ "Comparison": 91.5,
400
+ "Compound-Overall": 89.57,
401
+ "Imagination": 89.71,
402
+ "Feature matching": 89.25,
403
+ "Grammar-Overall": 86.8,
404
+ "Pronoun Reference": 97.62,
405
+ "Consistency": 88.89,
406
+ "Negation": 75.35,
407
+ "Layout-Overall": 93.33,
408
+ "2D": 93.36,
409
+ "3D": 93.28,
410
+ "Logical Reasoning": 83.09,
411
+ "Text": 91.85
412
+ },
413
+ {
414
+ "model": "Z-Image",
415
+ "link": "https://arxiv.org/pdf/2511.22699",
416
+ "hf": "https://huggingface.co/Tongyi-MAI/Z-Image",
417
+ "open_source": true,
418
+ "release_date": "2026-1",
419
+ "Overall": 86.77,
420
+ "Style": 97.26,
421
+ "World Knowledge": 94.36,
422
+ "Attribute-Overall": 93.25,
423
+ "Quantity": 85.11,
424
+ "Expression": 87.08,
425
+ "Material": 94.95,
426
+ "Size": 90.15,
427
+ "Shape": 87.41,
428
+ "Color": 97.8,
429
+ "Action-Overall": 83.72,
430
+ "Hand": 81.73,
431
+ "Full body": 79.38,
432
+ "Animal": 83.33,
433
+ "Non Contact": 86.16,
434
+ "Contact": 77.87,
435
+ "State": 87.3,
436
+ "Relationship-Overall": 89.06,
437
+ "Composition": 88.52,
438
+ "Similarity": 84.94,
439
+ "Inclusion": 93.39,
440
+ "Comparison": 89.56,
441
+ "Compound-Overall": 86.52,
442
+ "Imagination": 86.12,
443
+ "Feature matching": 87.38,
444
+ "Grammar-Overall": 80.46,
445
+ "Pronoun Reference": 91.67,
446
+ "Consistency": 80.16,
447
+ "Negation": 70.77,
448
+ "Layout-Overall": 91.11,
449
+ "2D": 92.51,
450
+ "3D": 89.31,
451
+ "Logical Reasoning": 76.72,
452
+ "Text": 75.27
453
+ },
454
  {
455
  "model": "FLUX.2-dev",
456
  "link": "https://github.com/black-forest-labs/flux2",