💻 Ranking: Código

Engenharia de código: SWE-Bench, LiveCodeBench, SciCode, Terminal-Bench e HumanEval.

Provedor
Licença
Modelos: 392
#ModeloPodium ScoreSWE-Bench VerifiedSWE-Bench ProLiveCodeBenchSciCodeTerminal-Bench HardHumanEvalVelocidade
01
84.1
95%70%60.2%62.9%71 t/s
02
Claude Mythos Preview
Anthropic · preview
80.9
93.9%77.8%80 t/s
03
Claude Opus 5
Anthropic
70.4
70%55.7%60 t/s
4
Claude Mythos 5
Anthropic · preview
66.0
50 t/s
5
64.8
64.6%73.7%56.1%65.9%72 t/s
6
64.4
63.3%58.2%208 t/s
7
Kimi K3
Moonshot AI · Aberto
61.5
74.7%58.7%37 t/s
8
60.2
69.7%52.7%233 t/s
9
58.0
88.6%69.2%67.7%53.5%58.3%50 t/s
10
GLM-5.1
Z.ai · Aberto
57.0
74 t/s
11
57.0
50 t/s
12
55.5
74%53.2%53%118 t/s
13
55.0
40 t/s
14
54.9
63.4%74%53.9%57.6%127 t/s
15
GPT-5.5
OpenAI
54.0
58.6%74.3%56.1%60.6%60 t/s
16
50.3
67.7%55 t/s
17
49.6
69.3%53.1%40.9%267 t/s
18
49.2
87.6%64.3%80 t/s
19
45.6
62.7%74%52.5%175 t/s
20
45.1
64.7%67.7%54.1%59 t/s
21
45.0
60 t/s
22
44.9
85.2%58.7%48.6%96 t/s
23
43.4
76.2%91.7%56.1%41.7%120 t/s
24
43.3
80.6%54.2%72.7%59%53.8%136 t/s
25
43.0
60 t/s
26
MiMo V2.5 Pro
Xiaomi · Aberto
42.6
78.9%73.3%50.2%43.2%65 t/s
27
GLM-5.2
Z.ai · Aberto
41.9
62.1%71.3%50.5%50.8%164 t/s
28
41.0
80 t/s
29
40.9
80.4%60.6%69%53.5%50.8%203 t/s
30
40.0
80 t/s
31
40.0
80 t/s
32
40.0
80 t/s
33
40.0
80 t/s
34
40.0
80 t/s
35
Qwen3 8 2 4t A95b
Alibaba · Aberto
40.0
85 t/s
36
40.0
90 t/s
37
40.0
80 t/s
38
40.0
90 t/s
39
Glm 5
Z.ai · Aberto
40.0
60 t/s
40
40.0
80 t/s
41
40.0
90 t/s
42
40.0
90 t/s
43
40.0
90 t/s
44
Kimi K3 Low
Moonshot AI
40.0
60 t/s
45
40.0
80 t/s
46
40.0
90 t/s
47
40.0
90 t/s
48
Deepseek V4 Pro 0424
DeepSeek · Aberto
40.0
60 t/s
49
40.0
80 t/s
50
40.0
90 t/s
51
40.0
50 t/s
52
Deepseek V4 Pro 0424 High
DeepSeek · Aberto
40.0
60 t/s
53
40.0
50 t/s
54
40.0
50 t/s
55
40.0
60 t/s
56
Solar Pro4
Upstage
40.0
40 t/s
57
40.0
90 t/s
58
40.0
90 t/s
59
Qwen3 6 Max
Alibaba · Aberto
40.0
85 t/s
60
40.0
80 t/s
61
40.0
90 t/s
62
40.0
60 t/s
63
40.0
90 t/s
64
40.0
90 t/s
65
40.0
150 t/s
66
Deepseek V3 2 Reasoning
DeepSeek · Aberto
40.0
40 t/s
67
40.0
80 t/s
68
Mimo V2 5 0424
Xiaomi · Aberto
40.0
60 t/s
69
Qwen3 6 27b
Alibaba · Aberto
40.0
85 t/s
70
40.0
50 t/s
71
40.0
80 t/s
72
40.0
90 t/s
73
40.0
80 t/s
74
Glm 5 1 Non Reasoning
Z.ai · Aberto
40.0
40 t/s
75
40.0
80 t/s
76
Kimi K2 5
Moonshot AI
40.0
60 t/s
77
40.0
60 t/s
78
40.0
90 t/s
79
40.0
40 t/s
80
40.0
50 t/s
81
40.0
80 t/s
82
Glm 5 2 Non Reasoning
Z.ai · Aberto
40.0
40 t/s
83
40.0
60 t/s
84
40.0
90 t/s
85
40.0
50 t/s
86
40.0
60 t/s
87
40.0
90 t/s
88
40.0
90 t/s
89
Kimi K2 Thinking
Moonshot AI
40.0
40 t/s
90
O3 Pro
OpenAI
40.0
60 t/s
91
Glm 5 Non Reasoning
Z.ai · Aberto
40.0
40 t/s
92
40.0
50 t/s
93
40.0
55 t/s
94
Qwen3 6 35b A3b
Alibaba · Aberto
40.0
85 t/s
95
40.0
40 t/s
96
40.0
150 t/s
97
40.0
40 t/s
98
Gpt 5 Low
OpenAI
40.0
90 t/s
99
40.0
50 t/s
100
Qwen3 5 Omni Plus
Alibaba · Aberto
40.0
85 t/s
101
Qwen3 6 27b Non Reasoning
Alibaba · Aberto
40.0
55 t/s
102
40.0
60 t/s
103
40.0
40 t/s
104
40.0
40 t/s
105
Qwen3 5 27b Non Reasoning
Alibaba · Aberto
40.0
55 t/s
106
40.0
80 t/s
107
40.0
50 t/s
108
Glm 4 6 Reasoning
Z.ai · Aberto
40.0
40 t/s
109
40.0
90 t/s
110
40.0
80 t/s
111
40.0
40 t/s
112
40.0
60 t/s
113
40.0
55 t/s
114
40.0
50 t/s
115
Glm 4 7 Non Reasoning
Z.ai · Aberto
40.0
40 t/s
116
40.0
60 t/s
117
40.0
40 t/s
118
40.0
60 t/s
119
40.0
60 t/s
120
Gemma 4 26b A4b
Google · Aberto
40.0
120 t/s
121
Claude 4 Opus
Anthropic
40.0
80 t/s
122
40.0
80 t/s
123
40.0
40 t/s
124
40.0
55 t/s
125
40.0
50 t/s
126
40.0
55 t/s
127
40.0
60 t/s
128
40.0
55 t/s
129
Kimi K2 0905
Moonshot AI
40.0
60 t/s
130
40.0
80 t/s
131
O1
OpenAI
40.0
60 t/s
132
40.0
150 t/s
133
40.0
50 t/s
134
40.0
60 t/s
135
40.0
80 t/s
136
Gemma 4 12b
Google · Aberto
40.0
120 t/s
137
40.0
50 t/s
138
Mercury 2
Inception
40.0
60 t/s
139
Qwen3 5 9b
Alibaba · Aberto
40.0
85 t/s
140
Deepseek V3 2 0925
DeepSeek · Aberto
40.0
60 t/s
141
40.0
60 t/s
142
40.0
40 t/s
143
Qwen3 Coder Next
Alibaba · Aberto
40.0
85 t/s
144
Deepseek V3 1 Reasoning
DeepSeek · Aberto
40.0
40 t/s
145
40.0
55 t/s
146
40.0
60 t/s
147
40.0
60 t/s
148
Qwen3 5 9b Non Reasoning
Alibaba · Aberto
40.0
55 t/s
149
40.0
80 t/s
150
Qwen3 5 4b
Alibaba · Aberto
40.0
85 t/s
151
40.0
55 t/s
152
40.0
40 t/s
153
Kimi K2
Moonshot AI
40.0
60 t/s
154
Mistral Small 4
Mistral · Aberto
40.0
100 t/s
155
Gpt 4 1
OpenAI
40.0
90 t/s
156
Qwen3 Max Preview
Alibaba · Aberto
40.0
85 t/s
157
Devstral 2
Mistral
40.0
100 t/s
158
O1 Pro
OpenAI
40.0
60 t/s
159
Deepseek R1 0120
DeepSeek · Aberto
40.0
40 t/s
160
40.0
80 t/s
161
40.0
60 t/s
162
40.0
85 t/s
163
40.0
40 t/s
164
40.0
85 t/s
165
Qwen3 Vl 32b Reasoning
Alibaba · Aberto
40.0
55 t/s
166
40.0
100 t/s
167
40.0
60 t/s
168
Glm 4 6v Reasoning
Z.ai · Aberto
40.0
40 t/s
169
40.0
55 t/s
170
40.0
40 t/s
171
40.0
80 t/s
172
Qwen3 5 4b Non Reasoning
Alibaba · Aberto
40.0
55 t/s
173
Mistral Large 3
Mistral · Aberto
40.0
100 t/s
174
Deepseek V3 0324
DeepSeek · Aberto
40.0
60 t/s
175
40.0
90 t/s
176
Mistral Small 3 1
Mistral · Aberto
40.0
100 t/s
177
Mistral Medium 3 1
Mistral · Aberto
40.0
100 t/s
178
40.0
55 t/s
179
Llama 4 Maverick
Meta · Aberto
40.0
80 t/s
180
40.0
90 t/s
181
40.0
85 t/s
182
Deepseek V3
DeepSeek · Aberto
40.0
60 t/s
183
40.0
85 t/s
184
40.0
85 t/s
185
40.0
55 t/s
186
40.0
55 t/s
187
Qwq 32b
Alibaba
40.0
85 t/s
188
40.0
80 t/s
189
40.0
60 t/s
190
Mistral Medium 3
Mistral · Aberto
40.0
100 t/s
191
Celeris 1
Celeris
40.0
60 t/s
192
40.0
50 t/s
193
40.0
60 t/s
194
Gemma 4 E4b
Google · Aberto
40.0
120 t/s
195
Claude 3 Opus
Anthropic
40.0
80 t/s
196
40.0
55 t/s
197
Gpt 4o
OpenAI
40.0
90 t/s
198
Qwen3 Vl 32b Instruct
Alibaba · Aberto
40.0
85 t/s
199
Qwen3 235b A22b Instruct
Alibaba · Aberto
40.0
85 t/s
200
Mistral Small 3 2
Mistral · Aberto
40.0
100 t/s
201
40.0
60 t/s
202
Llama 4 Scout
Meta · Aberto
40.0
80 t/s
203
40.0
50 t/s
204
Qwen3 Vl 30b A3b Instruct
Alibaba · Aberto
40.0
85 t/s
205
40.0
80 t/s
206
40.0
40 t/s
207
40.0
55 t/s
208
40.0
90 t/s
209
40.0
80 t/s
210
40.0
55 t/s
211
Glm 4 5v Reasoning
Z.ai · Aberto
40.0
40 t/s
212
40.0
60 t/s
213
40.0
50 t/s
214
Qwen3 30b A3b 2507
Alibaba · Aberto
40.0
85 t/s
215
40.0
50 t/s
216
40.0
80 t/s
217
40.0
80 t/s
218
40.0
80 t/s
219
Qwen3 32b Instruct
Alibaba · Aberto
40.0
85 t/s
220
40.0
90 t/s
221
40.0
80 t/s
222
40.0
90 t/s
223
Command A
Cohere
40.0
60 t/s
224
Nova Pro
Amazon
40.0
60 t/s
225
Gemma 3 27b
Google · Aberto
40.0
120 t/s
226
40.0
80 t/s
227
Mistral Large 2407
Mistral · Aberto
40.0
100 t/s
228
Gpt 4
OpenAI
40.0
90 t/s
229
40.0
80 t/s
230
Mistral Small 3
Mistral · Aberto
40.0
100 t/s
231
Qwen3 30b A3b Instruct
Alibaba · Aberto
40.0
85 t/s
232
40.0
80 t/s
233
Qwen Turbo
Alibaba · Aberto
40.0
85 t/s
234
Gemma 3 12b
Google · Aberto
40.0
120 t/s
235
Granite 4 0 H Small
IBM · Aberto
40.0
60 t/s
236
40.0
80 t/s
237
40.0
60 t/s
238
40.0
85 t/s
239
Phi 4
Microsoft · Aberto
40.0
60 t/s
240
40.0
60 t/s
241
Mistral Small
Mistral · Aberto
40.0
100 t/s
242
Mistral Large
Mistral · Aberto
40.0
100 t/s
243
40.0
80 t/s
244
40.0
90 t/s
245
Mistral Medium
Mistral · Aberto
40.0
100 t/s
246
Mistral Small 2402
Mistral · Aberto
40.0
100 t/s
247
Llama 3 Instruct 70b
Meta · Aberto
40.0
80 t/s
248
40.0
80 t/s
249
40.0
60 t/s
250
40.0
60 t/s
251
40.0
100 t/s
252
40.0
60 t/s
253
Gemma 3 4b
Google · Aberto
40.0
120 t/s
254
Gemma 3n E2b
Google · Aberto
40.0
120 t/s
255
Gemma 3n E4b
Google · Aberto
40.0
120 t/s
256
40.0
60 t/s
257
40.0
60 t/s
258
40.0
60 t/s
259
40.0
60 t/s
260
40.0
80 t/s
261
40.0
80 t/s
262
40.0
80 t/s
263
40.0
80 t/s
264
40.0
40 t/s
265
Falcon 2
TII · Aberto
40.0
60 t/s
266
40.0
60 t/s
267
40.0
90 t/s
268
40.0
90 t/s
269
40.0
80 t/s
270
40.0
60 t/s
271
Mistral
Mistral · Aberto
40.0
100 t/s
272
Nvidia
NVIDIA
40.0
150 t/s
273
40.0
150 t/s
274
40.0
60 t/s
275
Qwen Image
Alibaba · Aberto
40.0
85 t/s
276
Qwen3 Asr
Alibaba · Aberto
40.0
85 t/s
277
Kimi K2.7 Code
Moonshot AI · Aberto
39.9
66.3%47.5%44.7%39 t/s
278
Nova 2.0 Pro Preview
Amazon · preview
39.0
40 t/s
279
MiniMax M3
MiniMax
38.7
80.5%74%45.4%42.4%93 t/s
280
DeepSeek V4 Flash
DeepSeek · Aberto
38.0
79%52.6%91.6%49.9%113 t/s
281
GPT-oss-120B
OpenAI · Aberto
37.0
240 t/s
282
Kimi K2.6
Moonshot AI · Aberto
36.4
80.2%58.6%52.2%40 t/s
283
35.6
80.9%50 t/s
284
DeepSeek V3.2
DeepSeek · Aberto
35.5
73.1%83.3%60 t/s
285
Inkling
Thinking Machines
33.8
63.3%46.1%81 t/s
286
DeepSeek V4 Pro
DeepSeek · Aberto
30.7
80.6%55.4%93.5%42.4%36.4%66 t/s
287
30.3
77.7%57.6%51.3%54 t/s
288
Grok 4 Heavy
xAI · preview
30.0
79.4%50 t/s
289
28.0
40 t/s
290
GPT-oss-20B
OpenAI · Aberto
27.0
90 t/s
291
27.0
40 t/s
292
26.8
78%57.9%66.7%47.6%69 t/s
293
26.5
80.8%58.3%45.7%48.5%50 t/s
294
25.0
120 t/s
295
25.0
120 t/s
296
25.0
120 t/s
297
25.0
90 t/s
298
Deepseek V4 Flash 0420
DeepSeek · Aberto
25.0
60 t/s
299
25.0
90 t/s
300
Deepseek V4 Flash 0420 High
DeepSeek · Aberto
25.0
60 t/s
301
25.0
80 t/s
302
25.0
60 t/s
303
25.0
120 t/s
304
25.0
120 t/s
305
Minimax M2 1
MiniMax · Aberto
25.0
60 t/s
306
Mimo V2 Flash Reasoning
Xiaomi · Aberto
25.0
40 t/s
307
25.0
90 t/s
308
25.0
90 t/s
309
25.0
60 t/s
310
25.0
90 t/s
311
25.0
90 t/s
312
25.0
40 t/s
313
Minimax M2
MiniMax · Aberto
25.0
60 t/s
314
O4 Mini
OpenAI
25.0
60 t/s
315
25.0
60 t/s
316
25.0
90 t/s
317
25.0
50 t/s
318
25.0
120 t/s
319
25.0
40 t/s
320
25.0
80 t/s
321
25.0
60 t/s
322
25.0
90 t/s
323
25.0
90 t/s
324
25.0
40 t/s
325
O3 Mini
OpenAI
25.0
60 t/s
326
Qwen3 5 Omni Flash
Alibaba · Aberto
25.0
85 t/s
327
25.0
40 t/s
328
Minimax M1 80k
MiniMax · Aberto
25.0
60 t/s
329
25.0
60 t/s
330
25.0
90 t/s
331
25.0
60 t/s
332
25.0
60 t/s
333
25.0
40 t/s
334
25.0
80 t/s
335
25.0
150 t/s
336
25.0
90 t/s
337
25.0
100 t/s
338
25.0
90 t/s
339
25.0
60 t/s
340
25.0
120 t/s
341
25.0
60 t/s
342
25.0
80 t/s
343
25.0
100 t/s
344
25.0
120 t/s
345
Qwen3 Vl 8b Reasoning
Alibaba · Aberto
25.0
55 t/s
346
25.0
55 t/s
347
25.0
90 t/s
348
25.0
100 t/s
349
25.0
100 t/s
350
25.0
100 t/s
351
25.0
55 t/s
352
Qwen3 Vl 8b Instruct
Alibaba · Aberto
25.0
85 t/s
353
25.0
90 t/s
354
Llama 3 1 Instruct 8b
Meta · Aberto
25.0
80 t/s
355
25.0
150 t/s
356
25.0
150 t/s
357
25.0
100 t/s
358
Qwen3 14b Instruct
Alibaba · Aberto
25.0
85 t/s
359
25.0
120 t/s
360
25.0
90 t/s
361
Nova Lite
Amazon
25.0
60 t/s
362
25.0
60 t/s
363
Solar Mini
Upstage
25.0
60 t/s
364
Phi 4 Mini
Microsoft · Aberto
25.0
60 t/s
365
Qwen3 8b Instruct
Alibaba · Aberto
25.0
85 t/s
366
25.0
150 t/s
367
Llama 2 Chat 7b
Meta · Aberto
25.0
80 t/s
368
25.0
60 t/s
369
Olmo 3 7b Instruct
AI2 · Aberto
25.0
60 t/s
370
25.0
60 t/s
371
Mistral 7b Instruct
Mistral · Aberto
25.0
100 t/s
372
25.0
60 t/s
373
25.0
60 t/s
374
Gemma 3 1b
Google · Aberto
25.0
120 t/s
375
Llama 3 Instruct 8b
Meta · Aberto
25.0
80 t/s
376
25.0
60 t/s
377
Gemini
Google
25.0
120 t/s
378
25.0
120 t/s
379
25.0
120 t/s
380
Minimax H3
MiniMax · Aberto
25.0
60 t/s
381
24.8
77.4%52.4%69.7%51.5%45.5%80 t/s
382
GPT-5.4
OpenAI
23.4
57.7%60 t/s
383
22.6
79.6%50 t/s
384
GPT-5.2
OpenAI
21.0
80%90 t/s
385
20.4
78.8%56.6%69.7%40.7%43.9%55 t/s
386
Nemotron 3 Ultra 550B
NVIDIA · Aberto
19.1
67%39.9%36.4%142 t/s
387
14.8
78%120 t/s
388
GPT-5.1
OpenAI
14.6
76.3%60 t/s
389
Qwen3.5 397B-A17B
Alibaba · Aberto
11.5
76.4%71 t/s
390
Seed 2.0 Pro
ByteDance
9.9
76.5%40 t/s
391
GPT-5
OpenAI
8.2
74.9%93.4%60 t/s
392
6.4
74.5%80 t/s
O Podium Score é uma mistura ponderada de cinco leaderboards. Saiba mais na metodologia.