🧠 درجہ بندی: استدلال

گہرا استدلال اور سائنس: GPQA Diamond، HLE اور ARC-AGI-2۔

فراہم کنندہ
لائسنس
ماڈلز: 391
#ماڈلPodium ScoreGPQA DiamondHumanity’s Last ExamARC-AGI-2رفتار
01
Claude Mythos Preview
Anthropic · preview
98.8
94.6%64.7%80 t/s
02
87.6
94.6%47.2%72 t/s
03
Kimi K3
Moonshot AI · اوپن
86.5
93.5%56%37 t/s
4
82.8
93.6%57.9%50 t/s
5
Claude Opus 5
Anthropic
82.3
93.2%52.6%60 t/s
6
81.0
92.6%53.3%71 t/s
7
GPT-5.5
OpenAI
80.4
93.6%52.2%85%60 t/s
8
73.5
94.3%51.4%77.1%136 t/s
9
73.3
94.2%54.7%80 t/s
10
Claude Mythos 5
Anthropic · preview
71.0
50 t/s
11
69.6
92.9%41.8%127 t/s
12
66.5
92.2%41%267 t/s
13
66.3
92.6%43.6%55 t/s
14
66.0
91.3%53.1%68.8%50 t/s
15
65.7
92.8%38.3%233 t/s
16
GLM-5.2
Z.ai · اوپن
65.6
91.2%54.7%164 t/s
17
MiniMax M3
MiniMax
64.8
92.9%37.1%93 t/s
18
62.9
91.5%39.9%118 t/s
19
62.7
89.8%45.1%208 t/s
20
GLM-5.1
Z.ai · اوپن
62.0
74 t/s
21
62.0
50 t/s
22
61.0
92.4%41.4%203 t/s
23
60.8
93%40.3%59 t/s
24
GPT-5.4
OpenAI
60.1
92.8%39.8%73.3%60 t/s
25
60.0
40 t/s
26
GPT-5.1
OpenAI
55.5
88.1%60 t/s
27
55.1
92.3%37.2%175 t/s
28
55.0
50 t/s
29
55.0
50 t/s
30
55.0
50 t/s
31
55.0
60 t/s
32
Solar Pro4
Upstage
55.0
40 t/s
33
Deepseek V3 2 Reasoning
DeepSeek · اوپن
55.0
40 t/s
34
55.0
50 t/s
35
Glm 5 1 Non Reasoning
Z.ai · اوپن
55.0
40 t/s
36
55.0
60 t/s
37
55.0
40 t/s
38
55.0
50 t/s
39
Glm 5 2 Non Reasoning
Z.ai · اوپن
55.0
40 t/s
40
55.0
60 t/s
41
55.0
50 t/s
42
Kimi K2 Thinking
Moonshot AI
55.0
40 t/s
43
O3 Pro
OpenAI
55.0
60 t/s
44
Glm 5 Non Reasoning
Z.ai · اوپن
55.0
40 t/s
45
55.0
50 t/s
46
55.0
55 t/s
47
55.0
40 t/s
48
55.0
40 t/s
49
55.0
50 t/s
50
Qwen3 6 27b Non Reasoning
Alibaba · اوپن
55.0
55 t/s
51
55.0
40 t/s
52
55.0
40 t/s
53
Qwen3 5 27b Non Reasoning
Alibaba · اوپن
55.0
55 t/s
54
55.0
50 t/s
55
Glm 4 6 Reasoning
Z.ai · اوپن
55.0
40 t/s
56
Mimo V2 5 Pro Non Reasoning
Xiaomi · اوپن
55.0
40 t/s
57
55.0
60 t/s
58
55.0
55 t/s
59
55.0
50 t/s
60
Glm 4 7 Non Reasoning
Z.ai · اوپن
55.0
40 t/s
61
55.0
60 t/s
62
55.0
40 t/s
63
55.0
60 t/s
64
Deepseek V3 2 Reasoning 0925
DeepSeek · اوپن
55.0
40 t/s
65
Qwen3 Max Thinking Preview
Alibaba · اوپن
55.0
55 t/s
66
55.0
50 t/s
67
55.0
55 t/s
68
55.0
55 t/s
69
O1
OpenAI
55.0
60 t/s
70
55.0
50 t/s
71
Gemma 4 31b Non Reasoning
Google · اوپن
55.0
80 t/s
72
55.0
50 t/s
73
55.0
40 t/s
74
Deepseek V3 1 Reasoning
DeepSeek · اوپن
55.0
40 t/s
75
Qwen3 Vl 235b A22b Reasoning
Alibaba · اوپن
55.0
55 t/s
76
55.0
60 t/s
77
Qwen3 5 9b Non Reasoning
Alibaba · اوپن
55.0
55 t/s
78
55.0
80 t/s
79
55.0
55 t/s
80
55.0
40 t/s
81
O1 Pro
OpenAI
55.0
60 t/s
82
Deepseek R1 0120
DeepSeek · اوپن
55.0
40 t/s
83
55.0
40 t/s
84
Qwen3 Vl 32b Reasoning
Alibaba · اوپن
55.0
55 t/s
85
55.0
60 t/s
86
Glm 4 6v Reasoning
Z.ai · اوپن
55.0
40 t/s
87
Qwen3 Next 80b A3b Reasoning
Alibaba · اوپن
55.0
55 t/s
88
55.0
40 t/s
89
Qwen3 5 4b Non Reasoning
Alibaba · اوپن
55.0
55 t/s
90
Qwen3 30b A3b 2507 Reasoning
Alibaba · اوپن
55.0
55 t/s
91
55.0
55 t/s
92
Qwen3 Vl 30b A3b Reasoning
Alibaba · اوپن
55.0
55 t/s
93
Gemma 4 12b Non Reasoning
Google · اوپن
55.0
80 t/s
94
55.0
50 t/s
95
55.0
60 t/s
96
Qwen3 32b Instruct Reasoning
Alibaba · اوپن
55.0
55 t/s
97
55.0
50 t/s
98
Deepseek R1 Distill Llama 70b
DeepSeek · اوپن
55.0
40 t/s
99
Qwen3 Omni 30b A3b Reasoning
Alibaba · اوپن
55.0
55 t/s
100
55.0
55 t/s
101
Glm 4 5v Reasoning
Z.ai · اوپن
55.0
40 t/s
102
55.0
50 t/s
103
55.0
50 t/s
104
Gemma 4 E4b Non Reasoning
Google · اوپن
55.0
80 t/s
105
55.0
40 t/s
106
55.0
60 t/s
107
DeepSeek V4 Pro
DeepSeek · اوپن
54.5
90.1%48.2%66 t/s
108
Kimi K2.7 Code
Moonshot AI · اوپن
48.9
89.6%32.8%39 t/s
109
Kimi K2.6
Moonshot AI · اوپن
48.7
90.5%36.4%40 t/s
110
47.9
87%50 t/s
111
47.3
89.5%58.4%42.5%80 t/s
112
47.0
60 t/s
113
46.7
89.9%49%58.3%50 t/s
114
45.6
90.3%34.7%54 t/s
115
Qwen3.5 122B-A10B
Alibaba · اوپن
45.2
86.6%126 t/s
116
45.0
80 t/s
117
45.0
60 t/s
118
Grok 4 Heavy
xAI · preview
44.9
88.4%50.7%50 t/s
119
GPT-5.2
OpenAI
44.3
92.4%34.5%52.9%90 t/s
120
Nova 2.0 Pro Preview
Amazon · preview
43.0
40 t/s
121
42.7
90.4%31.6%69 t/s
122
DeepSeek V4 Flash
DeepSeek · اوپن
41.2
88.1%45.1%113 t/s
123
GPT-oss-120B
OpenAI · اوپن
40.0
240 t/s
124
MiMo V2.5 Pro
Xiaomi · اوپن
39.7
86.6%33.8%65 t/s
125
GPT-5
OpenAI
39.0
85.7%60 t/s
126
Inkling
Thinking Machines
37.3
87.2%29.7%81 t/s
127
36.9
90.4%28.8%55 t/s
128
36.2
91.9%45.8%31.1%120 t/s
129
35.0
80 t/s
130
35.0
80 t/s
131
35.0
80 t/s
132
35.0
80 t/s
133
35.0
80 t/s
134
Qwen3 8 2 4t A95b
Alibaba · اوپن
35.0
85 t/s
135
35.0
90 t/s
136
35.0
80 t/s
137
35.0
90 t/s
138
Glm 5
Z.ai · اوپن
35.0
60 t/s
139
35.0
80 t/s
140
35.0
90 t/s
141
35.0
90 t/s
142
35.0
90 t/s
143
Kimi K3 Low
Moonshot AI
35.0
60 t/s
144
35.0
80 t/s
145
35.0
90 t/s
146
35.0
90 t/s
147
Deepseek V4 Pro 0424
DeepSeek · اوپن
35.0
60 t/s
148
35.0
80 t/s
149
35.0
90 t/s
150
Deepseek V4 Pro 0424 High
DeepSeek · اوپن
35.0
60 t/s
151
35.0
90 t/s
152
35.0
90 t/s
153
Qwen3 6 Max
Alibaba · اوپن
35.0
85 t/s
154
35.0
80 t/s
155
35.0
90 t/s
156
35.0
60 t/s
157
35.0
90 t/s
158
35.0
90 t/s
159
35.0
80 t/s
160
35.0
150 t/s
161
35.0
80 t/s
162
Mimo V2 5 0424
Xiaomi · اوپن
35.0
60 t/s
163
Qwen3 6 27b
Alibaba · اوپن
35.0
85 t/s
164
35.0
80 t/s
165
35.0
90 t/s
166
35.0
80 t/s
167
35.0
80 t/s
168
Kimi K2 5
Moonshot AI
35.0
60 t/s
169
35.0
90 t/s
170
35.0
80 t/s
171
35.0
90 t/s
172
35.0
60 t/s
173
35.0
90 t/s
174
35.0
90 t/s
175
Qwen3 6 35b A3b
Alibaba · اوپن
35.0
85 t/s
176
35.0
150 t/s
177
Gpt 5 Low
OpenAI
35.0
90 t/s
178
Mimo V2 Flash Reasoning
Xiaomi · اوپن
35.0
40 t/s
179
Qwen3 5 Omni Plus
Alibaba · اوپن
35.0
85 t/s
180
35.0
60 t/s
181
35.0
80 t/s
182
35.0
40 t/s
183
35.0
90 t/s
184
35.0
80 t/s
185
35.0
60 t/s
186
Gemma 4 26b A4b
Google · اوپن
35.0
120 t/s
187
O4 Mini
OpenAI
35.0
60 t/s
188
Claude 4 Opus
Anthropic
35.0
80 t/s
189
35.0
80 t/s
190
35.0
60 t/s
191
Kimi K2 0905
Moonshot AI
35.0
60 t/s
192
35.0
80 t/s
193
35.0
150 t/s
194
35.0
50 t/s
195
35.0
60 t/s
196
Gemma 4 12b
Google · اوپن
35.0
120 t/s
197
Mercury 2
Inception
35.0
60 t/s
198
Qwen3 5 9b
Alibaba · اوپن
35.0
85 t/s
199
Deepseek V3 2 0925
DeepSeek · اوپن
35.0
60 t/s
200
35.0
60 t/s
201
Qwen3 Coder Next
Alibaba · اوپن
35.0
85 t/s
202
35.0
60 t/s
203
35.0
40 t/s
204
Qwen3 5 4b
Alibaba · اوپن
35.0
85 t/s
205
35.0
80 t/s
206
Kimi K2
Moonshot AI
35.0
60 t/s
207
Mistral Small 4
Mistral · اوپن
35.0
100 t/s
208
Gpt 4 1
OpenAI
35.0
90 t/s
209
Qwen3 Max Preview
Alibaba · اوپن
35.0
85 t/s
210
Devstral 2
Mistral
35.0
100 t/s
211
35.0
40 t/s
212
O3 Mini
OpenAI
35.0
60 t/s
213
35.0
80 t/s
214
35.0
60 t/s
215
35.0
85 t/s
216
35.0
85 t/s
217
35.0
40 t/s
218
35.0
60 t/s
219
35.0
100 t/s
220
35.0
60 t/s
221
35.0
80 t/s
222
Mistral Large 3
Mistral · اوپن
35.0
100 t/s
223
35.0
60 t/s
224
35.0
40 t/s
225
Deepseek V3 0324
DeepSeek · اوپن
35.0
60 t/s
226
35.0
80 t/s
227
35.0
90 t/s
228
Mistral Small 3 1
Mistral · اوپن
35.0
100 t/s
229
Mistral Medium 3 1
Mistral · اوپن
35.0
100 t/s
230
Llama 4 Maverick
Meta · اوپن
35.0
80 t/s
231
35.0
100 t/s
232
35.0
90 t/s
233
Qwen3 Vl 235b A22b Instruct
Alibaba · اوپن
35.0
85 t/s
234
Deepseek V3
DeepSeek · اوپن
35.0
60 t/s
235
Qwen3 Next 80b A3b Instruct
Alibaba · اوپن
35.0
85 t/s
236
Qwen3 Coder 30b A3b Instruct
Alibaba · اوپن
35.0
85 t/s
237
Qwq 32b
Alibaba
35.0
85 t/s
238
35.0
60 t/s
239
Mistral Medium 3
Mistral · اوپن
35.0
100 t/s
240
Celeris 1
Celeris
35.0
60 t/s
241
Gemma 4 E4b
Google · اوپن
35.0
120 t/s
242
Claude 3 Opus
Anthropic
35.0
80 t/s
243
35.0
80 t/s
244
Gpt 4o
OpenAI
35.0
90 t/s
245
Qwen3 Vl 32b Instruct
Alibaba · اوپن
35.0
85 t/s
246
Qwen3 235b A22b Instruct
Alibaba · اوپن
35.0
85 t/s
247
Mistral Small 3 2
Mistral · اوپن
35.0
100 t/s
248
Qwen3 Vl 8b Reasoning
Alibaba · اوپن
35.0
55 t/s
249
35.0
60 t/s
250
Qwen3 14b Instruct Reasoning
Alibaba · اوپن
35.0
55 t/s
251
Llama 4 Scout
Meta · اوپن
35.0
80 t/s
252
Qwen3 Vl 30b A3b Instruct
Alibaba · اوپن
35.0
85 t/s
253
35.0
80 t/s
254
35.0
90 t/s
255
Llama 3 3 Instruct 70b
Meta · اوپن
35.0
80 t/s
256
35.0
60 t/s
257
Qwen3 30b A3b 2507
Alibaba · اوپن
35.0
85 t/s
258
35.0
100 t/s
259
35.0
100 t/s
260
Hermes 4 Llama 3 1 405b
Meta · اوپن
35.0
80 t/s
261
35.0
80 t/s
262
Qwen3 32b Instruct
Alibaba · اوپن
35.0
85 t/s
263
35.0
90 t/s
264
Qwen3 8b Instruct Reasoning
Alibaba · اوپن
35.0
55 t/s
265
35.0
80 t/s
266
35.0
90 t/s
267
Command A
Cohere
35.0
60 t/s
268
Nova Pro
Amazon
35.0
60 t/s
269
Gemma 3 27b
Google · اوپن
35.0
120 t/s
270
35.0
80 t/s
271
Mistral Large 2407
Mistral · اوپن
35.0
100 t/s
272
Gpt 4
OpenAI
35.0
90 t/s
273
Hermes 4 Llama 3 1 70b
Meta · اوپن
35.0
80 t/s
274
Mistral Small 3
Mistral · اوپن
35.0
100 t/s
275
Qwen3 30b A3b Instruct
Alibaba · اوپن
35.0
85 t/s
276
Llama 3 1 Instruct 70b
Meta · اوپن
35.0
80 t/s
277
Qwen Turbo
Alibaba · اوپن
35.0
85 t/s
278
Gemma 3 12b
Google · اوپن
35.0
120 t/s
279
Granite 4 0 H Small
IBM · اوپن
35.0
60 t/s
280
Hermes 3 Llama 3 1 70b
Meta · اوپن
35.0
80 t/s
281
35.0
60 t/s
282
Qwen3 Omni 30b A3b Instruct
Alibaba · اوپن
35.0
85 t/s
283
Phi 4
Microsoft · اوپن
35.0
60 t/s
284
35.0
60 t/s
285
Mistral Small
Mistral · اوپن
35.0
100 t/s
286
Mistral Large
Mistral · اوپن
35.0
100 t/s
287
35.0
80 t/s
288
35.0
90 t/s
289
Mistral Medium
Mistral · اوپن
35.0
100 t/s
290
Mistral Small 2402
Mistral · اوپن
35.0
100 t/s
291
Llama 3 Instruct 70b
Meta · اوپن
35.0
80 t/s
292
35.0
80 t/s
293
35.0
60 t/s
294
35.0
60 t/s
295
35.0
100 t/s
296
35.0
60 t/s
297
Gemma 3 4b
Google · اوپن
35.0
120 t/s
298
Gemma 3n E2b
Google · اوپن
35.0
120 t/s
299
Gemma 3n E4b
Google · اوپن
35.0
120 t/s
300
35.0
60 t/s
301
35.0
60 t/s
302
35.0
60 t/s
303
35.0
60 t/s
304
35.0
80 t/s
305
35.0
80 t/s
306
35.0
80 t/s
307
35.0
80 t/s
308
Falcon 2
TII · اوپن
35.0
60 t/s
309
35.0
60 t/s
310
35.0
90 t/s
311
35.0
90 t/s
312
35.0
80 t/s
313
Mistral
Mistral · اوپن
35.0
100 t/s
314
Nvidia
NVIDIA
35.0
150 t/s
315
35.0
150 t/s
316
35.0
60 t/s
317
Qwen Image
Alibaba · اوپن
35.0
85 t/s
318
Qwen3 Asr
Alibaba · اوپن
35.0
85 t/s
319
Seed 2.0 Pro
ByteDance
34.2
88.9%40 t/s
320
Nemotron 3 Ultra 550B
NVIDIA · اوپن
33.4
87%26.6%142 t/s
321
32.7
90.4%43.5%33.6%120 t/s
322
31.0
40 t/s
323
Qwen3.5 397B-A17B
Alibaba · اوپن
29.0
88.4%28.7%71 t/s
324
GPT-oss-20B
OpenAI · اوپن
29.0
90 t/s
325
29.0
40 t/s
326
20.0
120 t/s
327
20.0
120 t/s
328
20.0
120 t/s
329
20.0
90 t/s
330
Deepseek V4 Flash 0420
DeepSeek · اوپن
20.0
60 t/s
331
20.0
90 t/s
332
Deepseek V4 Flash 0420 High
DeepSeek · اوپن
20.0
60 t/s
333
20.0
60 t/s
334
20.0
120 t/s
335
20.0
120 t/s
336
Minimax M2 1
MiniMax · اوپن
20.0
60 t/s
337
20.0
90 t/s
338
20.0
90 t/s
339
20.0
60 t/s
340
20.0
90 t/s
341
20.0
90 t/s
342
Minimax M2
MiniMax · اوپن
20.0
60 t/s
343
20.0
60 t/s
344
20.0
90 t/s
345
20.0
120 t/s
346
20.0
60 t/s
347
20.0
90 t/s
348
20.0
90 t/s
349
Qwen3 5 Omni Flash
Alibaba · اوپن
20.0
85 t/s
350
Minimax M1 80k
MiniMax · اوپن
20.0
60 t/s
351
20.0
90 t/s
352
20.0
150 t/s
353
20.0
90 t/s
354
20.0
90 t/s
355
20.0
60 t/s
356
20.0
120 t/s
357
20.0
60 t/s
358
20.0
100 t/s
359
20.0
120 t/s
360
20.0
90 t/s
361
20.0
100 t/s
362
Qwen3 Vl 8b Instruct
Alibaba · اوپن
20.0
85 t/s
363
20.0
90 t/s
364
Llama 3 1 Instruct 8b
Meta · اوپن
20.0
80 t/s
365
20.0
150 t/s
366
20.0
150 t/s
367
20.0
100 t/s
368
Qwen3 14b Instruct
Alibaba · اوپن
20.0
85 t/s
369
20.0
120 t/s
370
20.0
90 t/s
371
Nova Lite
Amazon
20.0
60 t/s
372
20.0
60 t/s
373
Solar Mini
Upstage
20.0
60 t/s
374
Phi 4 Mini
Microsoft · اوپن
20.0
60 t/s
375
Qwen3 8b Instruct
Alibaba · اوپن
20.0
85 t/s
376
20.0
150 t/s
377
Llama 2 Chat 7b
Meta · اوپن
20.0
80 t/s
378
20.0
60 t/s
379
Olmo 3 7b Instruct
AI2 · اوپن
20.0
60 t/s
380
20.0
60 t/s
381
Mistral 7b Instruct
Mistral · اوپن
20.0
100 t/s
382
20.0
60 t/s
383
20.0
60 t/s
384
Gemma 3 1b
Google · اوپن
20.0
120 t/s
385
Llama 3 Instruct 8b
Meta · اوپن
20.0
80 t/s
386
20.0
60 t/s
387
Gemini
Google
20.0
120 t/s
388
20.0
120 t/s
389
20.0
120 t/s
390
Minimax H3
MiniMax · اوپن
20.0
60 t/s
391
0.0
80%17.8%96 t/s
Podium Score پانچ لیڈر بورڈز کا وزنی امتزاج ہے۔ مزید جانیں طریقہ کار میں۔