Melhor IA para Programação em 2026Atualizado por AA Coding Index

Em 2026, Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback) lideram este ranking de 558 modelos para programação. A classificação usa o AA Coding Index como sinal principal, com fallback para LiveCodeBench e SciCode.

Sincronizado: 28 de outubro de 2026 • 558 modelos com benchmarks de código

Casos de Uso

Autocompletar Código

Sugestões inline enquanto você digita. Ideal para IDEs como Cursor e VS Code.

Top modelos: Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback)

Geração de Código

Criar funções, classes e projetos completos a partir de descrições em linguagem natural.

Top modelos: Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback)

Debug e Code Review

Identificar bugs, sugerir correções e revisar pull requests automaticamente.

Top modelos: Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback)

Ranking das Melhores IAs para Programação

#ModeloEmpresaScore de CodingBenchmarkContextPreço InputOpen Source
🥇Claude Fable 5.1 (Max, Default Fallback)AnthropicAnthropic
81.6
AA Coding Index1.0M tokens$10.00—
🥈Claude Fable 5.1 (Xhigh, Default Fallback)AnthropicAnthropic
80.7
AA Coding Index—$10.00—
🥉Claude Fable 5.1 (High, Default Fallback)AnthropicAnthropic
79.1
AA Coding Index—$10.00—
4GPT-5.6 Sol (Xhigh)OpenAIOpenAI
78.3
AA Coding Index—$4.00—
5Claude Opus 5 (Max)AnthropicAnthropic
78.0
AA Coding Index1.0M tokens$5.00—
6GPT-5.6 Sol (Max)OpenAIOpenAI
77.4
AA Coding Index1.1M tokens$4.00—
7GPT-5.6 Sol (High)OpenAIOpenAI
77.2
AA Coding Index—$4.00—
8GPT-6 Astra (High)OpenAIOpenAI
77.1
AA Coding Index—$10.00—
9Claude Fable 5.1 (Medium, Default Fallback)AnthropicAnthropic
77.1
AA Coding Index—$10.00—
10Claude Opus 5 (Xhigh)AnthropicAnthropic
77.0
AA Coding Index—$5.00—
11GPT-6 Astra (Max)OpenAIOpenAI
76.9
AA Coding Index1.1M tokens$10.00—
12Grok 4.6 (High)SpaceXAISpaceXAI
76.8
AA Coding Index500K tokens$2.00—
13GPT-5.6 Terra (Max)OpenAIOpenAI
76.7
AA Coding Index1.1M tokens$2.00—
14GPT-6 Astra (Medium)OpenAIOpenAI
76.7
AA Coding Index—$10.00—
15Claude Fable 5 (Max, Opus 4.8 Fallback)AnthropicAnthropic
76.5
AA Coding Index1.0M tokens$10.00—
16Muse Spark 1.3 (Xhigh)MetaMeta
76.5
AA Coding Index—$1.25—
17Claude Opus 5 (High)AnthropicAnthropic
76.5
AA Coding Index1.0M tokens$5.00—
18GPT-5.6 Sol (Medium)OpenAIOpenAI
76.3
AA Coding Index—$4.00—
19Gemini 3.8 Flash (High)GoogleGoogle
76.3
AA Coding Index1.0M tokens$0.75—
20Qwen3.8 Max (0902)AlibabaAlibaba
76.2
AA Coding Index1.0M tokens$2.00✅
21Kimi K3 (Max)KimiKimi
76.2
AA Coding Index1.0M tokens$3.00—
22Gemini 3.7 Flash (High)GoogleGoogle
76.1
AA Coding Index1.0M tokens$0.75—
23GPT-6 Astra (Xhigh)OpenAIOpenAI
75.9
AA Coding Index—$10.00—
24Grok 4.6 (Xhigh)SpaceXAISpaceXAI
75.9
AA Coding Index—$2.00—
25Muse Spark 1.3 (Max)MetaMeta
75.8
AA Coding Index—$1.25—
26GPT-6 Astra (Low)OpenAIOpenAI
75.7
AA Coding Index—$10.00—
27Claude Fable 5.1 (Low, Default Fallback)AnthropicAnthropic
75.2
AA Coding Index—$10.00—
28GPT-5.5 (Xhigh)OpenAIOpenAI
74.9
AA Coding Index1.1M tokens$5.00—
29GLM-5.3 (Max)Z AI
74.8
AA Coding Index—$1.40—
30Grok 4.6 (Medium)SpaceXAISpaceXAI
74.4
AA Coding Index—$2.00—
31Claude Opus 4.8 (Max)AnthropicAnthropic
74.3
AA Coding Index1.0M tokens$5.00—
32Claude Opus 5 (Medium)AnthropicAnthropic
74.3
AA Coding Index1.0M tokens$5.00—
33Gemini 3.8 Flash (Medium)GoogleGoogle
74.1
AA Coding Index—$0.75—
34Claude Opus 4.7 (Max)AnthropicAnthropic
73.6
AA Coding Index1.0M tokens$5.00—
35Gemini 3.8 Flash (Low)GoogleGoogle
73.5
AA Coding Index—$0.75—
36Qwen3.8-Flash-NextAlibabaAlibaba
73.1
AA Coding Index—$0.15—
37Grok 4.5 (High)SpaceXAISpaceXAI
72.4
AA Coding Index500K tokens$2.00—
38Muse Spark 1.2 (Xhigh)MetaMeta
72.2
AA Coding Index—$1.25—
39Kimi K3 (Low)KimiKimi
72.0
AA Coding Index1.0M tokens$3.00—
40Qwen3.8 2.4T A95BAlibabaAlibaba
71.9
AA Coding Index1.0M tokens$2.00✅
41Qwen3.8 MaxAlibabaAlibaba
71.8
AA Coding Index—$2.00—
42GPT-5.5 (High)OpenAIOpenAI
71.6
AA Coding Index—$5.00—
43Claude Sonnet 5 (Max)AnthropicAnthropic
71.5
AA Coding Index1.0M tokens$2.00—
44GLM-5.3-FlashZ AI
71.5
AA Coding Index—$0.15—
45Gemini 3.7 Flash (Medium)GoogleGoogle
71.5
AA Coding Index———
46GPT-5.5 (Medium)OpenAIOpenAI
71.5
AA Coding Index—$5.00—
47GPT-5.6 Luna (Max)OpenAIOpenAI
71.4
AA Coding Index1.1M tokens$0.20—
48Muse Spark 1.1 (Xhigh)MetaMeta
71.3
AA Coding Index—$1.25—
49GPT-5.4 (Xhigh)OpenAIOpenAI
71.1
AA Coding Index1.1M tokens$2.50—
50Gemini 3.7 Flash (Low)GoogleGoogle
71.0
AA Coding Index———
51GPT-5.6 Terra (Xhigh)OpenAIOpenAI
70.6
AA Coding Index—$2.00—
52Gemini 3.5 Flash (High)GoogleGoogle
70.1
AA Coding Index1.0M tokens$1.50—
53GPT-5.6 Sol (Low)OpenAIOpenAI
69.7
AA Coding Index—$4.00—
54Gemini 3.6 Flash (High)GoogleGoogle
69.2
AA Coding Index1.0M tokens$0.75—
55DeepSeek V4 Flash 0731 (Max)DeepSeekDeepSeek
69.1
AA Coding Index1.0M tokens$0.14—
56Gemini 3.1 Pro PreviewGoogleGoogle
68.8
AA Coding Index1.0M tokens$2.00—
57DeepSeek V4 Pro 0813 (Max)DeepSeekDeepSeek
68.8
AA Coding Index1.0M tokens$1.32—
58GLM-5.2 (Max)Z AI
68.8
AA Coding Index—$1.40—
59GPT-5.6 Luna (Xhigh)OpenAIOpenAI
68.6
AA Coding Index—$0.20—
60Qwen3.8 27B (Xhigh)AlibabaAlibaba
68.1
AA Coding Index—$0.50—
61GPT-5.6 Terra (High)OpenAIOpenAI
67.1
AA Coding Index—$2.00—
62Claude Opus 5 (Low)AnthropicAnthropic
66.9
AA Coding Index—$5.00—
63Claude Sonnet 5 (Non-reasoning, High)AnthropicAnthropic
66.4
AA Coding Index1.0M tokens$2.00—
64Grok 4.6 (Low)SpaceXAISpaceXAI
66.3
AA Coding Index—$2.00—
65Qwen3.7 MaxAlibabaAlibaba
66.0
AA Coding Index—$2.50—
66GPT-5.6 Sol (Non-reasoning)OpenAIOpenAI
65.1
AA Coding Index—$4.00—
67DeepSeek V4 Flash Vision (Max)DeepSeekDeepSeek
65.0
AA Coding Index—$0.44—
68GPT-5.6 Terra (Medium)OpenAIOpenAI
64.7
AA Coding Index—$2.00—
69Motif 3Motif Technologies
63.5
AA Coding Index———
70GPT-5.6 Luna (High)OpenAIOpenAI
63.3
AA Coding Index—$0.20—
71Claude Sonnet 4.6 (Max)AnthropicAnthropic
63.0
AA Coding Index—$3.00—
72Agnes 2.5 Pro BetaSapiens AI
62.3
AA Coding Index—$0.10—
73Motif 3 (Beta)Motif Technologies
62.0
AA Coding Index———
74Kimi K2.6 (Reasoning)KimiKimi
61.8
AA Coding Index—$0.95—
75K2 Horizon 375B A23BInstitute of Foundation Models
61.5
AA Coding Index———
76Quasar 438B (Max, Based on GLM-5.2)Multiverse Computing
61.2
AA Coding Index—$0.11—
77GPT-5.5 (Low)OpenAIOpenAI
60.9
AA Coding Index—$5.00—
78Kimi K2.7 CodeKimiKimi
60.8
AA Coding Index—$0.95—
79Apodex 1.1Apodex
60.8
AA Coding Index—$0.30—
80MiMo-V2.5-Pro (Reasoning)Xiaomi
60.2
AA Coding Index1.0M tokens$0.41—
81KAT Coder Pro V2KwaiKAT
59.5
AA Coding Index256K tokens——
82DeepSeek V4 Pro 0424 (Max)DeepSeekDeepSeek
59.4
AA Coding Index—$0.43—
83Nex-N2-Pro (Based on Qwen3.5-397B-A17B)Nex AGI
59.1
AA Coding Index262K tokens——
84Hy3Tencent
58.8
AA Coding Index262K tokens$0.14—
85DeepSeek V4 Pro 0424 (High)DeepSeekDeepSeek
58.7
AA Coding Index—$0.43—
86Muse SparkMetaMeta
58.6
AA Coding Index———
87MiniMax-M3MiniMax
58.6
AA Coding Index1.0M tokens$0.30—
88Qwen3.8 27B (Low)AlibabaAlibaba
58.2
AA Coding Index—$0.50—
89GPT-5.6 Terra (Low)OpenAIOpenAI
58.1
AA Coding Index—$2.00—
90Ling-3.0-flash-VLInclusionAI
57.0
AA Coding Index———
91MiMo-V2.5Xiaomi
56.8
AA Coding Index—$0.13—
92GPT-5.5 (Non-reasoning)OpenAIOpenAI
56.5
AA Coding Index—$5.00—
93DeepSeek V4 Flash 0420 (Max)DeepSeekDeepSeek
56.2
AA Coding Index—$0.11—
94GPT-5.4 mini (Xhigh)OpenAIOpenAI
56.1
AA Coding Index400K tokens$0.75—
95GPT-5.4 nano (Xhigh)OpenAIOpenAI
56.1
AA Coding Index400K tokens$0.20—
96Qwen3.8 27B (Medium)AlibabaAlibaba
56.1
AA Coding Index—$0.50—
97Qwen3.7 PlusAlibabaAlibaba
55.9
AA Coding Index—$0.40—
98GLM-5.1 (Reasoning)Z AI
55.8
AA Coding Index203K tokens$1.28✅
99Ling-3.0-flash-FinInclusionAI
55.6
AA Coding Index———
100Qwen3.6 PlusAlibabaAlibaba
54.5
AA Coding Index1.0M tokens$0.50✅
101Qwen3.6 27B (Reasoning)AlibabaAlibaba
53.7
AA Coding Index—$0.60—
102Inkling SmallThinking Machines
52.9
AA Coding Index—$0.30—
103MiniMax-M2.7MiniMax
52.6
AA Coding Index197K tokens$0.30✅
104JT-4.1 Flash 236B A21B (Non-reasoning)China Mobile
52.4
AA Coding Index———
105GPT-5.6 Terra (Non-reasoning)OpenAIOpenAI
52.3
AA Coding Index—$2.00—
106Claude 4.5 Sonnet (Reasoning)AnthropicAnthropic
52.1
AA Coding Index—$3.00—
107Inkling (Xhigh)Thinking Machines
52.1
AA Coding Index—$1.00—
108DeepSeek V4 Flash 0420 (High)DeepSeekDeepSeek
52.0
AA Coding Index—$0.11—
109Grok Build 0.1 0616SpaceXAISpaceXAI
51.5
AA Coding Index—$1.00—
110GPT-5.6 Luna (Medium)OpenAIOpenAI
50.7
AA Coding Index—$0.20—
111Ling 3.0 FlashInclusionAI
50.6
AA Coding Index—$0.07—
112MiMo-V2-Flash (Non-reasoning)Xiaomi
49.8
AA Coding Index———
113GPT-5.1 (High)OpenAIOpenAI
49.4
AA Coding Index400K tokens$1.25—
114Gemini 3.5 Flash-LiteGoogleGoogle
49.3
AA Coding Index1.0M tokens$0.30—
115Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIANVIDIA
49.3
AA Coding Index1.0M tokens$0.60—
116Muse Glimmer (High)MetaMeta
49.0
AA Coding Index—$0.30—
117Qwen3.5 397B A17B (Reasoning)AlibabaAlibaba
48.2
AA Coding Index262K tokens$0.60✅
118Mistral Medium 3.5MistralMistral
46.9
AA Coding Index262K tokens$1.50—
119Kimi K2.5 (Reasoning)KimiKimi
46.8
AA Coding Index262K tokens$0.60✅
120Gemini 2.5 Pro Preview (Mar' 25)GoogleGoogle
46.7
AA Coding Index———
121Qwen3.6 27B (Non-reasoning)AlibabaAlibaba
46.6
AA Coding Index—$0.60—
122GLM-5.2 (Non-reasoning)Z AI
46.5
AA Coding Index—$1.40—
123GLM-4.6 (Reasoning)Z AI
45.8
AA Coding Index—$0.55—
124Qwen3.5 122B A10B (Reasoning)AlibabaAlibaba
45.7
AA Coding Index262K tokens$0.40✅
125LongCat 2.0LongCat
45.3
AA Coding Index—$0.75—
126GLM-4.7 (Reasoning)Z AI
45.3
AA Coding Index—$0.53—
127Solar Open2 250BUpstage
45.0
AA Coding Index———
128Qwen3.8 27B (Non-reasoning)AlibabaAlibaba
44.6
AA Coding Index—$0.50—
129GPT-5.6 Luna (Low)OpenAIOpenAI
44.2
AA Coding Index—$0.20—
130DeepSeek V3.2 (Reasoning)DeepSeekDeepSeek
44.2
AA Coding Index—$0.28—
131Claude 4.5 Haiku (Reasoning)AnthropicAnthropic
43.9
AA Coding Index—$1.00—
132DeepSeek V3.1 Terminus (Reasoning)DeepSeekDeepSeek
43.5
AA Coding Index—$1.64—
133Gemma 4 31B (Reasoning)GoogleGoogle
43.4
AA Coding Index262K tokens——
134Qwen3.5 122B A10B (Non-reasoning)AlibabaAlibaba
43.3
AA Coding Index—$0.40—
135Ring-2.6-1TInclusionAI
42.8
AA Coding Index—$0.30—
136Grok 4.3 (High)SpaceXAISpaceXAI
42.2
AA Coding Index1.0M tokens$1.25—
137Qwen3.6 35B A3B (Reasoning)AlibabaAlibaba
41.9
AA Coding Index262K tokens$0.38✅
138K-EXAONE 2.0 0803LG AI Research
40.6
AA Coding Index———
139o1OpenAIOpenAI
39.7
AA Coding Index200K tokens$15.00—
140Step 3.7 FlashStepFun
39.6
AA Coding Index—$0.20—
141GPT-5.5 Instant (June 2026)OpenAIOpenAI
39.4
AA Coding Index—$5.00—
142GPT-5.6 Luna (Non-reasoning)OpenAIOpenAI
39.3
AA Coding Index—$0.20—
143Gemma 4 26B A4B (Reasoning)GoogleGoogle
39.3
AA Coding Index262K tokens$0.10—
144A.X-K2SK Telecom
38.8
AA Coding Index———
145K2 Horizon 7BInstitute of Foundation Models
38.6
AA Coding Index———
146GPT-5 (High)OpenAIOpenAI
37.8
AA Coding Index400K tokens$1.25—
147Nemotron 3 Super 120B A12B (Reasoning)NVIDIANVIDIA
37.7
AA Coding Index1.0M tokens$0.30—
148Claude 4 Sonnet (Reasoning)AnthropicAnthropic
37.6
AA Coding Index———
149Qwen3.5 35B A3B (Non-reasoning)AlibabaAlibaba
37.0
AA Coding Index—$0.25—
150North Mini CodeCohereCohere
36.5
AA Coding Index———
151Claude 3.7 Sonnet (Reasoning)AnthropicAnthropic
36.4
AA Coding Index200K tokens——
152Qwen3 Coder NextAlibabaAlibaba
36.2
AA Coding Index262K tokens$0.31✅
153Grok 4.3 (Non-reasoning)SpaceXAISpaceXAI
35.2
AA Coding Index—$1.25—
154Gemini 3.1 Flash-LiteGoogleGoogle
34.7
AA Coding Index1.0M tokens$0.25—
155Nova 2.0 Pro Preview (Medium)AmazonAmazon
34.0
AA Coding Index—$1.25—
156o1-previewOpenAIOpenAI
34.0
AA Coding Index—$16.50—
157Gemini 2.5 ProGoogleGoogle
33.3
AA Coding Index1.0M tokens$1.25—
158Gemma 4 31B (Non-reasoning)GoogleGoogle
33.2
AA Coding Index—$0.17—
159G9v3-39A5BAI9Stars
33.1
AA Coding Index———
160K-EXAONE (Reasoning)LG AI Research
32.1
AA Coding Index———
161Devstral 2MistralMistral
31.3
AA Coding Index———
162Mercury 2Inception
31.1
AA Coding Index128K tokens$0.25—
163Gemma 4 12B (Reasoning)GoogleGoogle
31.0
AA Coding Index—$0.10—
164gpt-oss-120b (High)OpenAIOpenAI
30.4
AA Coding Index131K tokens$0.15—
165Claude 3.5 Sonnet (Oct '24)AnthropicAnthropic
30.2
AA Coding Index—$3.00—
166Granite 4.2 30BIBM
29.9
AA Coding Index———
167Devstral Small 2MistralMistral
29.3
AA Coding Index———
168Qwen3.5 9B (Reasoning)AlibabaAlibaba
28.7
AA Coding Index—$0.14—
169Qwen3.6 35B A3B (Non-reasoning)AlibabaAlibaba
28.1
AA Coding Index—$0.38—
170Command A+CohereCohere
27.8
AA Coding Index192K tokens——
171Nemotron 3.5 LightningNVIDIANVIDIA
26.8
AA Coding Index262K tokens$0.06—
172Mistral Small 4 (Reasoning)MistralMistral
26.6
AA Coding Index—$0.15—
173Ling 3.0 TinyInclusionAI
26.5
AA Coding Index———
174Mistral Small 3.1MistralMistral
26.3
AA Coding Index—$0.08—
175K2 Horizon 3.7BInstitute of Foundation Models
26.1
AA Coding Index———
176Claude 3.5 Sonnet (June '24)AnthropicAnthropic
26.0
AA Coding Index—$3.00—
177Nova 2.0 Pro Preview (Low)AmazonAmazon
25.9
AA Coding Index—$1.25—
178Trinity Large ThinkingArcee AI
25.8
AA Coding Index262K tokens—✅
179Gemini 2.0 Pro Experimental (Feb '25)GoogleGoogle
25.5
AA Coding Index———
180Nemotron Cascade 2 30B A3BNVIDIANVIDIA
25.3
AA Coding Index———
181Ling 2.6 FlashInclusionAI
25.3
AA Coding Index———
182DeepSeek R1 (Jan '25)DeepSeekDeepSeek
24.6
AA Coding Index—$2.00—
183GPT-4o (May '24)OpenAIOpenAI
24.2
AA Coding Index128K tokens$5.00—
184Gemini 2.0 Flash Thinking Experimental (Jan '25)GoogleGoogle
24.1
AA Coding Index———
185Gemini 1.5 Pro (Sep '24)GoogleGoogle
23.6
AA Coding Index———
186EXAONE 4.5 33B (Reasoning)LG AI Research
23.6
AA Coding Index———
187Qwen3.5 9B (Non-reasoning)AlibabaAlibaba
23.5
AA Coding Index—$0.17—
188HyperNova 60B 2605 (High, Based on gpt-oss-120b)Multiverse Computing
23.2
AA Coding Index———
189Nova 2.0 Lite (High)AmazonAmazon
23.0
AA Coding Index—$0.30—
190DeepSeek V3 (Dec '24)DeepSeekDeepSeek
23.0
AA Coding Index164K tokens$0.32✅
191Qwen3.5 4B (Reasoning)AlibabaAlibaba
22.6
AA Coding Index—$0.03—
192Granite 4.2 8BIBM
22.4
AA Coding Index—$0.06—
193Qwen3 235B A22B 2507 (Reasoning)AlibabaAlibaba
22.1
AA Coding Index—$0.23—
194GPT-4 TurboOpenAIOpenAI
21.5
AA Coding Index128K tokens$10.00—
195Magistral Medium 1.2MistralMistral
21.3
AA Coding Index———
196DeepSeek V3 0324DeepSeekDeepSeek
21.2
AA Coding Index—$0.24—
197gpt-oss-120b (Low)OpenAIOpenAI
21.2
AA Coding Index131K tokens$0.13—
198K2 Think V2Institute of Foundation Models
21.0
AA Coding Index———
199Nova 2.0 Pro Preview (Non-reasoning)AmazonAmazon
20.9
AA Coding Index—$1.25—
200gpt-oss-20b (High)OpenAIOpenAI
20.7
AA Coding Index131K tokens$0.05—
201Mistral Medium 3.1MistralMistral
20.5
AA Coding Index131K tokens—✅
202Qwen3.5 4B (Non-reasoning)AlibabaAlibaba
20.3
AA Coding Index—$0.03—
203GPT-4.1 miniOpenAIOpenAI
20.2
AA Coding Index1.0M tokens$0.40—
204Mistral Large 3MistralMistral
20.1
AA Coding Index—$0.50—
205Gemini 1.5 Pro (May '24)GoogleGoogle
19.8
AA Coding Index———
206DiffusionGemma 26B A4BGoogleGoogle
19.7
AA Coding Index———
207Claude 3 OpusAnthropicAnthropic
19.5
AA Coding Index—$15.00—
208Gemini 1.0 UltraGoogleGoogle
17.6
AA Coding Index———
209Granite 4.2 3BIBM
17.5
AA Coding Index—$0.03—
210Qwen3 Next 80B A3B (Reasoning)AlibabaAlibaba
17.4
AA Coding Index———
211o3-mini (High)OpenAIOpenAI
16.3
AA Coding Index200K tokens$1.10—
212Llama 4 MaverickMetaMeta
16.3
AA Coding Index1.0M tokens$0.25✅
213Solar Pro 3Upstage
16.2
AA Coding Index128K tokens$0.15—
214Claude 3.5 HaikuAnthropicAnthropic
15.9
AA Coding Index200K tokens——
215GPT-5 mini (High)OpenAIOpenAI
15.6
AA Coding Index400K tokens$0.25—
216Qwen3 32B (Reasoning)AlibabaAlibaba
15.3
AA Coding Index—$0.24—
217Magistral Small 1.2MistralMistral
14.7
AA Coding Index—$0.50—
218MiniCPM5-2BOpenBMB
14.5
AA Coding Index———
219NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIANVIDIA
14.4
AA Coding Index—$0.05—
220Ministral 3 14BMistralMistral
14.4
AA Coding Index—$0.20—
221Celeris-1Celeris
14.4
AA Coding Index—$0.20—
222Claude 2.1AnthropicAnthropic
14.0
AA Coding Index———
223Qwen3 14B (Reasoning)AlibabaAlibaba
13.8
AA Coding Index—$0.12—
224Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIANVIDIA
13.8
AA Coding Index—$0.40—
225GPT-4OpenAIOpenAI
13.1
AA Coding Index8K tokens——
226Claude 2.0AnthropicAnthropic
12.9
AA Coding Index———
227Mistral Small 3.2MistralMistral
12.5
AA Coding Index—$0.07—
228Qwen3 30B A3B 2507 (Reasoning)AlibabaAlibaba
12.1
AA Coding Index———
229Llama 3.3 Instruct 70BMetaMeta
11.9
AA Coding Index131K tokens$0.60✅
230GPT-4o miniOpenAIOpenAI
11.4
AA Coding Index128K tokens$0.15—
231GPT-4.1 nanoOpenAIOpenAI
11.1
AA Coding Index1.0M tokens$0.10—
232GPT-3.5 TurboOpenAIOpenAI
10.7
AA Coding Index———
233Granite 4.1 30BIBM
10.4
AA Coding Index———
234Gemma 3 27B InstructGoogleGoogle
10.1
AA Coding Index131K tokens——
235G9v3-3BAI9Stars
9.9
AA Coding Index———
236Ministral 3 8BMistralMistral
9.7
AA Coding Index—$0.15—
237Nanbeige4.1-3BNanbeige
9.6
AA Coding Index———
238Granite 4.1 8BIBM
9.5
AA Coding Index———
239Qwen3 8B (Reasoning)AlibabaAlibaba
9.0
AA Coding Index———
240Llama 4 ScoutMetaMeta
8.2
AA Coding Index1.3M tokens$0.18✅
241NVIDIA Nemotron 3 Nano 4BNVIDIANVIDIA
8.0
AA Coding Index———
242Claude InstantAnthropicAnthropic
7.8
AA Coding Index———
243LFM2.5-2.6BLiquid AI
7.7
AA Coding Index———
244Gemma 4 E2B (Reasoning)GoogleGoogle
7.2
AA Coding Index———
245Gemma 3 12B InstructGoogleGoogle
5.8
AA Coding Index131K tokens——
246Llama 3.1 Instruct 8BMetaMeta
5.4
AA Coding Index16K tokens$0.04✅
247Ministral 3 3BMistralMistral
4.8
AA Coding Index—$0.10—
248Granite 4.1 3BIBM
4.7
AA Coding Index———
249PALM-2GoogleGoogle
4.6
AA Coding Index———
250Phi-4 Mini InstructMicrosoftMicrosoft
3.8
AA Coding Index—$0.07—
251K2 Horizon 0.9BInstitute of Foundation Models
3.4
AA Coding Index———
252Gemma 3n E4B InstructGoogleGoogle
3.2
AA Coding Index———
253Qwen3.5 2B (Reasoning)AlibabaAlibaba
2.9
AA Coding Index———
254Gemma 3 4B InstructGoogleGoogle
2.7
AA Coding Index131K tokens——
255Qwen3.5 2B (Non-reasoning)AlibabaAlibaba
2.4
AA Coding Index———
256Qwen3.5 0.8B (Non-reasoning)AlibabaAlibaba
1.2
AA Coding Index———
257MiniCPM-V 4.6 1.3BOpenBMB
0.7
AA Coding Index———
258Gemini 3 Pro Preview (High)GoogleGoogle
92.0
LiveCodeBench—$2.00—
259Gemini 3 Flash Preview (Reasoning)GoogleGoogle
91.0
LiveCodeBench—$0.50—
260DeepSeek V3.2 SpecialeDeepSeekDeepSeek
90.0
LiveCodeBench164K tokens—✅
261GPT-5.2 (Xhigh)OpenAIOpenAI
89.0
LiveCodeBench400K tokens$1.75—
262Claude Opus 4.5 (Reasoning)AnthropicAnthropic
87.0
LiveCodeBench—$5.00—
263MiMo-V2-Flash (Reasoning)Xiaomi
87.0
LiveCodeBench262K tokens$0.10—
264MiMo-V2-Flash (Feb 2026)Xiaomi
86.8
LiveCodeBench———
265Gemini 3 Pro Preview (Low)GoogleGoogle
86.0
LiveCodeBench—$2.00—
266o4-mini (High)OpenAIOpenAI
86.0
LiveCodeBench200K tokens$1.10—
267DeepSeek V3.2 Exp (Reasoning)DeepSeekDeepSeek
86.0
LiveCodeBench—$0.28—
268Kimi K2 ThinkingKimiKimi
85.0
LiveCodeBench262K tokens$0.60—
269GPT-5.1 Codex (High)OpenAIOpenAI
85.0
LiveCodeBench400K tokens$1.25—
270GPT-5.1 Codex mini (High)OpenAIOpenAI
84.0
LiveCodeBench400K tokens$0.25—
271GPT-5 Codex (High)OpenAIOpenAI
84.0
LiveCodeBench400K tokens$1.25—
272Grok 4 Fast (Non-reasoning)SpaceXAISpaceXAI
83.0
LiveCodeBench2.0M tokens$0.20—
273MiniMax-M2MiniMax
83.0
LiveCodeBench205K tokens$0.30—
274Grok 4SpaceXAISpaceXAI
82.0
LiveCodeBench256K tokens$3.00—
275Grok 4.1 Fast (Non-reasoning)SpaceXAISpaceXAI
82.0
LiveCodeBench2.0M tokens——
276MiniMax-M2.1MiniMax
81.0
LiveCodeBench197K tokens$0.30✅
277o3OpenAIOpenAI
81.0
LiveCodeBench200K tokens$2.00—
278ERNIE 5.0 Thinking PreviewBaidu
81.0
LiveCodeBench———
279Apriel-v1.6-15B-ThinkerServiceNow
81.0
LiveCodeBench———
280o3-proOpenAIOpenAI
80.8
LiveCodeBench200K tokens$20.00—
281Gemini 3 Flash Preview (Non-reasoning)GoogleGoogle
80.0
LiveCodeBench—$0.50—
282DeepSeek V3.1 Terminus (Non-reasoning)DeepSeekDeepSeek
80.0
LiveCodeBench164K tokens$0.27✅
283Qwen3 235B A22B 2507 InstructAlibabaAlibaba
79.0
LiveCodeBench262K tokens$0.21✅
284GPT-5 nano (High)OpenAIOpenAI
79.0
LiveCodeBench400K tokens$0.05—
285INTELLECT-3 (Based on GLM-4.5-Air)Prime Intellect
78.0
LiveCodeBench131K tokens—✅
286DeepSeek V3.1 (Non-reasoning)DeepSeekDeepSeek
78.0
LiveCodeBench164K tokens$0.56✅
287GPT-5.4 Pro (Xhigh)OpenAIOpenAI
77.5
LiveBench Coding1.1M tokens$30.00—
288Gemini 2.5 Pro Preview (May' 25)GoogleGoogle
77.0
LiveCodeBench—$1.25—
289Qwen3 MaxAlibabaAlibaba
77.0
LiveCodeBench262K tokens—✅
290Seed-OSS-36B-InstructByteDance Seed
77.0
LiveCodeBench—$0.21—
291Claude Sonnet 4.5AnthropicAnthropic
76.1
LiveBench Coding1.0M tokens$3.00—
292KAT-Coder-Pro V1KwaiKAT
75.0
LiveCodeBench———
293EXAONE 4.0 32B (Reasoning)LG AI Research
75.0
LiveCodeBench———
294Claude Opus 4.5 (Non-reasoning)AnthropicAnthropic
74.0
LiveCodeBench200K tokens$5.00—
295GLM-4.5 (Reasoning)Z AI
74.0
LiveCodeBench131K tokens——
296Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIANVIDIA
74.0
LiveCodeBench———
297Qwen3 VL 32B (Reasoning)AlibabaAlibaba
74.0
LiveCodeBench———
298Apriel-v1.5-15B-ThinkerServiceNow
73.0
LiveCodeBench———
299o3-miniOpenAIOpenAI
72.0
LiveCodeBench200K tokens$1.10—
300Falcon-H1R-7BTII UAE
72.0
LiveCodeBench———
301NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIANVIDIA
72.0
LiveCodeBench—$0.04—
302Gemini 2.5 Flash Preview (Sep '25) (Reasoning)GoogleGoogle
71.0
LiveCodeBench———
303MiniMax M1 80kMiniMax
71.0
LiveCodeBench—$0.55—
304NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIANVIDIA
70.1
LiveCodeBench131K tokens$0.05—
305Gemini 2.5 Flash (Reasoning)GoogleGoogle
70.0
LiveCodeBench—$0.30—
306Olmo 3.1 32B ThinkAllen Institute for AI
70.0
LiveCodeBench———
307Qwen3 VL 30B A3B (Reasoning)AlibabaAlibaba
70.0
LiveCodeBench—$0.50—
308Cogito v2.1 (Reasoning)Deep Cogito
69.0
LiveCodeBench———
309Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)GoogleGoogle
69.0
LiveCodeBench—$0.10—
310NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIANVIDIA
69.0
LiveCodeBench———
311Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research
69.0
LiveCodeBench—$1.00—
312Ling-1TInclusionAI
68.0
LiveCodeBench———
313Qwen3 Omni 30B A3B (Reasoning)AlibabaAlibaba
68.0
LiveCodeBench—$0.25—
314Qwen3 Next 80B A3B InstructAlibabaAlibaba
68.0
LiveCodeBench262K tokens$0.10✅
315Olmo 3 32B ThinkAllen Institute for AI
67.0
LiveCodeBench66K tokens—✅
316GPT-5.2 Codex (Xhigh)OpenAIOpenAI
66.9
LiveCodeBench400K tokens$1.75—
317MiniMax M1 40kMiniMax
66.0
LiveCodeBench———
318Nova 2.0 Omni (Medium)AmazonAmazon
66.0
LiveCodeBench———
319Grok Code Fast 1SpaceXAISpaceXAI
66.0
LiveCodeBench256K tokens——
320Mi:dm K 2.5 ProKorea Telecom
66.0
LiveCodeBench———
321Claude 4.1 Opus (Non-reasoning)AnthropicAnthropic
65.4
LiveCodeBench—$15.00—
322SpaceXAI: Grok Build 0.1xAIxAI
65.4
LiveBench Coding256K tokens$1.00—
323Claude 4.1 Opus (Reasoning)AnthropicAnthropic
65.0
LiveCodeBench—$15.00—
324Qwen3 VL 235B A22B (Reasoning)AlibabaAlibaba
65.0
LiveCodeBench———
325Qwen3 Max (Preview)AlibabaAlibaba
65.0
LiveCodeBench———
326Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research
65.0
LiveCodeBench———
327Motif-2-12.7B-ReasoningMotif Technologies
65.0
LiveCodeBench———
328Claude 4 Opus (Reasoning)AnthropicAnthropic
64.0
LiveCodeBench—$15.00—
329Ring-1TInclusionAI
64.0
LiveCodeBench———
330Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIANVIDIA
64.0
LiveCodeBench———
331Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)GoogleGoogle
64.0
LiveCodeBench—$0.10—
332Qwen3 4B 2507 (Reasoning)AlibabaAlibaba
64.0
LiveCodeBench———
333QwQ 32BAlibabaAlibaba
63.0
LiveCodeBench———
334HyperCLOVA X SEED Think (32B)Naver
63.0
LiveCodeBench———
335Ring-flash-2.0InclusionAI
63.0
LiveCodeBench———
336Qwen3 235B A22B (Reasoning)AlibabaAlibaba
62.0
LiveCodeBench———
337Solar Pro 2 (Non-reasoning)Upstage
62.0
LiveCodeBench———
338Olmo 3 7B ThinkAllen Institute for AI
62.0
LiveCodeBench———
339DeepSeek R1 0528 (May '25)DeepSeekDeepSeek
61.7
LiveCodeBench64K tokens$1.85✅
340Kimi K2 0905KimiKimi
61.0
LiveCodeBench262K tokens$0.60✅
341GLM-4.5V (Reasoning)Z AI
60.0
LiveCodeBench—$0.60—
342Claude 4.5 Sonnet (Non-reasoning)AnthropicAnthropic
59.0
LiveCodeBench—$3.00—
343DeepSeek V3.2 (Non-reasoning)DeepSeekDeepSeek
59.0
LiveCodeBench164K tokens$0.28✅
344Qwen3 VL 235B A22B InstructAlibabaAlibaba
59.0
LiveCodeBench262K tokens$0.21✅
345Qwen3 Coder 480B A35B InstructAlibabaAlibaba
59.0
LiveCodeBench—$0.38—
346Nova 2.0 Omni (Low)AmazonAmazon
59.0
LiveCodeBench———
347Ling-flash-2.0InclusionAI
59.0
LiveCodeBench—$0.14—
348Gemini 2.5 Flash-Lite (Non-reasoning)GoogleGoogle
59.0
LiveCodeBench1.0M tokens$0.10—
349o1-miniOpenAIOpenAI
58.0
LiveCodeBench———
350Mi:dm K 2.5 Pro PreviewKorea Telecom
58.0
LiveCodeBench———
351GPT-5 (Minimal)OpenAIOpenAI
56.0
LiveCodeBench—$1.25—
352Kimi K2KimiKimi
56.0
LiveCodeBench131K tokens$0.57—
353DeepSeek V3.2 Exp (Non-reasoning)DeepSeekDeepSeek
55.0
LiveCodeBench—$0.28—
354GPT-5 mini (Minimal)OpenAIOpenAI
55.0
LiveCodeBench—$0.25—
355Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research
55.0
LiveCodeBench—$1.00—
356Claude 4 Opus (Non-reasoning)AnthropicAnthropic
54.0
LiveCodeBench200K tokens$15.00—
357Qwen3 Max Thinking (Preview)AlibabaAlibaba
54.0
LiveCodeBench———
358GPT-5 (ChatGPT)OpenAIOpenAI
54.0
LiveCodeBench———
359K2-V2 (Medium)Institute of Foundation Models
54.0
LiveCodeBench———
360Magistral Medium 1MistralMistral
53.0
LiveCodeBench———
361GPT-5.3 Codex (Xhigh)OpenAIOpenAI
53.0
SciCode400K tokens$1.75—
362Qwen3 30B A3B 2507 InstructAlibabaAlibaba
52.0
LiveCodeBench—$0.10—
363Claude Opus 4.6 (Max)AnthropicAnthropic
52.0
SciCode—$5.00—
364Claude 4.5 Haiku (Non-reasoning)AnthropicAnthropic
51.0
LiveCodeBench200K tokens$1.00—
365Qwen3 VL 32B InstructAlibabaAlibaba
51.0
LiveCodeBench131K tokens—✅
366Qwen3 30B A3B (Reasoning)AlibabaAlibaba
51.0
LiveCodeBench—$0.14—
367Magistral Small 1MistralMistral
51.0
LiveCodeBench———
368DeepSeek R1 0528 Qwen3 8BDeepSeekDeepSeek
51.0
LiveCodeBench———
369Gemini 2.5 Flash (Non-reasoning)GoogleGoogle
50.0
LiveCodeBench1.0M tokens$0.30—
370GPT-5.5 Instant (May 2026)OpenAIOpenAI
50.0
SciCode—$5.00—
371Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIANVIDIA
49.0
LiveCodeBench———
372Gemini 3.5 Flash (Minimal)GoogleGoogle
49.0
SciCode—$1.50—
373Qwen3 VL 30B A3B InstructAlibabaAlibaba
48.0
LiveCodeBench131K tokens$0.35✅
374ERNIE 4.5 300B A47BBaidu
47.0
LiveCodeBench123K tokens—✅
375GPT-5 nano (Minimal)OpenAIOpenAI
47.0
LiveCodeBench—$0.05—
376EXAONE 4.0 32B (Non-reasoning)LG AI Research
47.0
LiveCodeBench———
377Qwen3 4B (Reasoning)AlibabaAlibaba
47.0
LiveCodeBench———
378Qwen3.6 Max PreviewAlibabaAlibaba
47.0
SciCode———
379Claude Sonnet 4.6 (Non-reasoning, High)AnthropicAnthropic
47.0
SciCode1.0M tokens$3.00—
380GPT-4.1OpenAIOpenAI
46.0
LiveCodeBench1.0M tokens$2.00—
381Solar Pro 2 (Preview) (Reasoning)Upstage
46.0
LiveCodeBench———
382Claude Opus 4.6 (Non-reasoning, High)AnthropicAnthropic
46.0
SciCode1.0M tokens$5.00—
383Claude 4 Sonnet (Non-reasoning)AnthropicAnthropic
45.0
LiveCodeBench200K tokens——
384Grok 4.20 0309 (Reasoning)SpaceXAISpaceXAI
45.0
SciCode—$2.00—
385Reka Flash (Sep '24)Reka AI
44.0
LiveCodeBench66K tokens$0.20✅
386GLM-5-TurboZ AI
44.0
SciCode203K tokens——
387Claude Sonnet 4.6 (Non-reasoning, Low)AnthropicAnthropic
44.0
SciCode—$3.00—
388GPT-4o (March 2025, chatgpt-4o-latest)OpenAIOpenAI
43.0
LiveCodeBench———
389Ling-mini-2.0InclusionAI
43.0
LiveCodeBench———
390MiMo-V2-ProXiaomi
43.0
SciCode1.0M tokens——
391MiniMax-M2.5MiniMax
43.0
SciCode197K tokens$0.30✅
392Qwen3 Max ThinkingAlibabaAlibaba
43.0
SciCode262K tokens—✅
393Qwen3 Omni 30B A3B InstructAlibabaAlibaba
42.0
LiveCodeBench—$0.25—
394Gemini 2.5 Flash Preview (Non-reasoning)GoogleGoogle
41.0
LiveCodeBench———
395Hy3-preview (Reasoning)Tencent
41.0
SciCode262K tokens$0.06—
396Qwen3.5 Omni PlusAlibabaAlibaba
41.0
SciCode—$0.40—
397Mistral Medium 3MistralMistral
40.0
LiveCodeBench131K tokens$0.40✅
398Qwen3 Coder 30B A3B InstructAlibabaAlibaba
40.0
LiveCodeBench160K tokens$0.19✅
399MiMo-V2-Omni-0327Xiaomi
40.0
SciCode———
400Qwen3.5 27B (Reasoning)AlibabaAlibaba
40.0
SciCode262K tokens$0.30✅
401Step 3.5 FlashStepFun
40.0
SciCode—$0.10—
402Claude 3.7 Sonnet (Non-reasoning)AnthropicAnthropic
39.0
LiveCodeBench200K tokens$3.00—
403Solar Pro 2 (Preview) (Non-reasoning)Upstage
39.0
LiveCodeBench———
404Step 3.5 Flash 2603StepFun
39.0
SciCode262K tokens$0.10✅
405DeepSeek R1 Distill Qwen 14BDeepSeekDeepSeek
38.0
LiveCodeBench———
406Kimi Linear 48B A3B InstructKimiKimi
38.0
LiveCodeBench———
407Qwen3 4B 2507 InstructAlibabaAlibaba
38.0
LiveCodeBench———
408JT-4.1 Flash 236B A21B (Reasoning)China Mobile
38.0
SciCode———
409Qwen3.5 35B A3B (Reasoning)AlibabaAlibaba
38.0
SciCode262K tokens$0.25✅
410GLM-5 (Reasoning)Z AI
38.0
SciCode205K tokens$1.00—
411Ling-2.6-1TInclusionAI
37.0
SciCode—$0.30—
412MiMo-V2-OmniXiaomi
37.0
SciCode262K tokens——
413Qwen2.5 MaxAlibabaAlibaba
36.0
LiveCodeBench———
414NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIANVIDIA
36.0
LiveCodeBench262K tokens$0.05—
415GLM-5.1 (Non-reasoning)Z AI
36.0
SciCode—$1.38—
416Qwen3 VL 8B (Reasoning)AlibabaAlibaba
35.0
LiveCodeBench———
417NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIANVIDIA
35.0
LiveCodeBench—$0.20—
418Qwen: Qwen3 235B A22B Thinking 2507AlibabaAlibaba
34.3
LiveCodeBench131K tokens$0.15✅
419Devstral MediumMistralMistral
34.0
LiveCodeBench131K tokens—✅
420QwQ 32B-PreviewAlibabaAlibaba
34.0
LiveCodeBench———
421Gemini 2.0 Flash (Feb '25)GoogleGoogle
33.0
LiveCodeBench1.0M tokens——
422Qwen3 VL 8B InstructAlibabaAlibaba
33.0
LiveCodeBench131K tokens—✅
423GPT-4o (ChatGPT)OpenAIOpenAI
33.0
SciCode———
424Qwen: Qwen3 30B A3B Thinking 2507AlibabaAlibaba
32.2
LiveCodeBench131K tokens$0.08✅
425GPT-4o (Aug '24)OpenAIOpenAI
32.0
LiveCodeBench128K tokens$2.50—
426Qwen3 30B A3B (Non-reasoning)AlibabaAlibaba
32.0
LiveCodeBench262K tokens$0.12✅
427Qwen3 VL 4B (Reasoning)AlibabaAlibaba
32.0
LiveCodeBench———
428GPT-4o (Nov '24)OpenAIOpenAI
31.0
LiveCodeBench128K tokens$2.50—
429Llama 3.1 Instruct 405BMetaMeta
31.0
LiveCodeBench———
430Nova 2.0 Omni (Non-reasoning)AmazonAmazon
31.0
LiveCodeBench———
431Qwen3 1.7B (Reasoning)AlibabaAlibaba
31.0
LiveCodeBench———
432Step3 VL 10BStepFun
31.0
SciCode———
433Qwen2.5 Coder Instruct 32BAlibabaAlibaba
30.0
LiveCodeBench33K tokens—✅
434SonarPerplexityPerplexity
30.0
LiveCodeBench127K tokens——
435Sarvam M (Reasoning, Based on Mistral Small 3.1)Sarvam
30.0
LiveCodeBench———
436Llama 3.1 Tulu3 405BAllen Institute for AI
29.0
LiveCodeBench———
437Mistral Large 2 (Nov '24)MistralMistral
29.0
LiveCodeBench———
438Qwen3 32B (Non-reasoning)AlibabaAlibaba
29.0
LiveCodeBench—$0.15—
439Command ACohereCohere
29.0
LiveCodeBench256K tokens$2.50—
440Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIANVIDIA
29.0
LiveCodeBench———
441Qwen3 VL 4B InstructAlibabaAlibaba
29.0
LiveCodeBench———
442JT-35B-FlashChina Mobile
29.0
SciCode———
443Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIANVIDIA
28.0
LiveCodeBench———
444Qwen3 14B (Non-reasoning)AlibabaAlibaba
28.0
LiveCodeBench—$0.12—
445Qwen2.5 Instruct 72BAlibabaAlibaba
28.0
LiveCodeBench33K tokens$0.47✅
446Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIANVIDIA
28.0
LiveCodeBench———
447Sonar Reasoning ProPerplexityPerplexity
28.0
LiveCodeBench128K tokens——
448LongCat Flash LiteLongCat
28.0
SciCode———
449DeepSeek R1 Distill Qwen 32BDeepSeekDeepSeek
27.0
LiveCodeBench128K tokens—✅
450DeepSeek R1 Distill Llama 70BDeepSeekDeepSeek
27.0
LiveCodeBench8K tokens$0.70✅
451Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research
27.0
LiveCodeBench———
452Grok 2 (Dec '24)SpaceXAISpaceXAI
27.0
LiveCodeBench———
453Gemini 1.5 Flash (Sep '24)GoogleGoogle
27.0
LiveCodeBench———
454Mistral Large 2 (Jul '24)MistralMistral
27.0
LiveCodeBench131K tokens$2.00—
455Olmo 3 7B InstructAllen Institute for AI
27.0
LiveCodeBench———
456JT-MINIChina Mobile
27.0
SciCode———
457Solar Open 100B (Reasoning)Upstage
27.0
SciCode———
458Pixtral LargeMistralMistral
26.0
LiveCodeBench131K tokens——
459Devstral Small (May '25)MistralMistral
26.0
LiveCodeBench———
460Qwen3.5 Omni FlashAlibabaAlibaba
26.0
SciCode—$0.10—
461Sarvam 105B (High)Sarvam
26.0
SciCode—$0.04—
462Devstral Small (Jul '25)MistralMistral
25.0
LiveCodeBench131K tokens——
463Mistral Small 3MistralMistral
25.0
LiveCodeBench—$0.05—
464Qwen2.5 Instruct 32BAlibabaAlibaba
25.0
LiveCodeBench———
465Granite 4.0 H SmallIBM
25.0
LiveCodeBench———
466Grok BetaSpaceXAISpaceXAI
24.0
LiveCodeBench———
467Gemma 4 E4B (Reasoning)GoogleGoogle
24.0
SciCode—$0.02—
468Mistral SabaMistralMistral
24.0
SciCode33K tokens—✅
469Llama 3.1 Instruct 70BMetaMeta
23.0
LiveCodeBench131K tokens$0.56✅
470Phi-4MicrosoftMicrosoft
23.0
LiveCodeBench16K tokens$0.13✅
471Qwen3 4B (Non-reasoning)AlibabaAlibaba
23.0
LiveCodeBench———
472DeepSeek R1 Distill Llama 8BDeepSeekDeepSeek
23.0
LiveCodeBench———
473Gemini 1.5 Flash-8BGoogleGoogle
22.0
LiveCodeBench———
474Gemini 2.0 Flash (experimental)GoogleGoogle
21.0
LiveCodeBench———
475Llama 3.2 Instruct 90B (Vision)MetaMeta
21.0
LiveCodeBench———
476Jamba Reasoning 3BAI21 Labs
21.0
LiveCodeBench———
477DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research
20.0
LiveCodeBench———
478Llama 3 Instruct 70BMetaMeta
20.0
LiveCodeBench8K tokens$0.65✅
479Gemini 1.5 Flash (May '24)GoogleGoogle
20.0
LiveCodeBench———
480Qwen3 8B (Non-reasoning)AlibabaAlibaba
20.0
LiveCodeBench—$0.20—
481Gemma 4 E2B (Non-reasoning)GoogleGoogle
20.0
SciCode———
482Gemini 2.0 Flash-Lite (Feb '25)GoogleGoogle
19.0
LiveCodeBench———
483Hermes 3 - Llama-3.1 70BNous Research
19.0
LiveCodeBench—$0.70—
484Sarvam 30B (High)Sarvam
19.0
SciCode———
485Gemini 2.0 Flash LiteGoogleGoogle
18.5
LiveCodeBench1.0M tokens$0.07—
486Gemini 2.0 Flash-Lite (Preview)GoogleGoogle
18.0
LiveCodeBench———
487Claude 3 SonnetAnthropicAnthropic
18.0
LiveCodeBench———
488Jamba 1.7 LargeAI21 Labs
18.0
LiveCodeBench256K tokens—✅
489Granite 4.0 MicroIBM
18.0
LiveCodeBench131K tokens—✅
490Tri-21B-think PreviewTrillion Labs
18.0
SciCode———
491Mistral Large (Feb '24)MistralMistral
17.8
LiveCodeBench128K tokens$4.00✅
492Llama 3.1 Nemotron Instruct 70BNVIDIANVIDIA
17.0
LiveCodeBench131K tokens—✅
493Jamba 1.6 LargeAI21 Labs
17.0
LiveCodeBench———
494Tri-21B-ThinkTrillion Labs
17.0
SciCode———
495Olmo 3.1 32B InstructAllen Institute for AI
17.0
SciCode66K tokens—✅
496GLM-4.6V (Reasoning)Z AI
16.0
LiveCodeBench—$0.30—
497Qwen2 Instruct 72BAlibabaAlibaba
16.0
LiveCodeBench———
498DeepSeek Coder V2 Lite InstructDeepSeekDeepSeek
16.0
LiveCodeBench———
499Mixtral 8x22B InstructMistralMistral
15.0
LiveCodeBench———
500Claude 3 HaikuAnthropicAnthropic
15.0
LiveCodeBench200K tokens$0.25—
501LFM2 8B A1BLiquid AI
15.0
LiveCodeBench———
502Mistral Small (Sep '24)MistralMistral
14.0
LiveCodeBench———
503Jamba 1.5 LargeAI21 Labs
14.0
LiveCodeBench—$2.00—
504Gemma 3n E4B Instruct Preview (May '25)GoogleGoogle
14.0
LiveCodeBench———
505Qwen2.5 Coder Instruct 7BAlibabaAlibaba
13.0
LiveCodeBench———
506Phi-4 Multimodal InstructMicrosoftMicrosoft
13.0
LiveCodeBench—$0.08—
507Granite 3.3 8B (Non-reasoning)IBM
13.0
LiveCodeBench———
508Qwen3 1.7B (Non-reasoning)AlibabaAlibaba
13.0
LiveCodeBench———
509Molmo2-8BAllen Institute for AI
13.0
SciCode———
510Cohere: Command R+ (08-2024)CohereCohere
12.2
LiveCodeBench128K tokens$2.50—
511Command-R+ (Apr '24)CohereCohere
12.0
LiveCodeBench———
512Gemini 1.0 ProGoogleGoogle
12.0
LiveCodeBench———
513Phi-3 Mini Instruct 3.8BMicrosoftMicrosoft
12.0
LiveCodeBench———
514Granite 4.0 H 1BIBM
12.0
LiveCodeBench———
515Qwen3 0.6B (Reasoning)AlibabaAlibaba
12.0
LiveCodeBench———
516OpenChat 3.5 (1210)OpenChat
12.0
LiveCodeBench———
517Mistral Small (Feb '24)MistralMistral
11.0
LiveCodeBench———
518Llama 3.2 Instruct 11B (Vision)MetaMeta
11.0
LiveCodeBench131K tokens$0.34✅
519LFM2 24B A2BLiquid AI
11.0
SciCode33K tokens—✅
520Llama 3 Instruct 8BMetaMeta
10.0
LiveCodeBench8K tokens$0.04✅
521Mistral MediumMistralMistral
10.0
LiveCodeBench———
522Llama 2 Chat 13BMetaMeta
10.0
LiveCodeBench———
523LFM 40BLiquid AI
10.0
LiveCodeBench———
524Gemma 3n E2B InstructGoogleGoogle
10.0
LiveCodeBench———
525Llama 2 Chat 70BMetaMeta
10.0
LiveCodeBench———
526DBRX InstructDatabricks
9.0
LiveCodeBench———
527DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research
9.0
LiveCodeBench———
528Llama 3.2 Instruct 3BMetaMeta
8.0
LiveCodeBench80K tokens—✅
529LFM2 2.6BLiquid AI
8.0
LiveCodeBench———
530LFM2.5-8B-A1BLiquid AI
8.0
SciCode———
531Jamba 1.6 MiniAI21 Labs
7.0
LiveCodeBench———
532OLMo 2 32BAllen Institute for AI
7.0
LiveCodeBench———
533DeepSeek R1 Distill Qwen 1.5BDeepSeekDeepSeek
7.0
LiveCodeBench———
534Qwen3 0.6B (Non-reasoning)AlibabaAlibaba
7.0
LiveCodeBench———
535Mixtral 8x7B InstructMistralMistral
7.0
LiveCodeBench33K tokens$0.45✅
536Jamba 1.7 MiniAI21 Labs
6.0
LiveCodeBench———
537Jamba 1.5 MiniAI21 Labs
6.0
LiveCodeBench—$0.20—
538Apertus 70B InstructSwiss AI Initiative
6.0
SciCode———
539Granite 4.0 1BIBM
5.0
LiveCodeBench———
540Command-R (Mar '24)CohereCohere
5.0
LiveCodeBench———
541Mistral 7B InstructMistralMistral
5.0
LiveCodeBench—$0.15—
542OLMo 2 7BAllen Institute for AI
4.0
LiveCodeBench———
543Molmo 7B-DAllen Institute for AI
4.0
LiveCodeBench———
544MiniCPM5-1B (Non-reasoning)OpenBMB
4.0
SciCode———
545Gemma 4 E4B (Non-reasoning)GoogleGoogle
4.0
SciCode—$0.02—
546Tiny Aya GlobalCohereCohere
4.0
SciCode———
547LFM2.5-1.2B-ThinkingLiquid AI
4.0
SciCode———
548Apertus 8B InstructSwiss AI Initiative
4.0
SciCode—$0.10—
549LFM2.5-VL-1.6BLiquid AI
3.0
SciCode———
550LFM2 1.2BLiquid AI
2.0
LiveCodeBench———
551Granite 4.0 H 350MIBM
2.0
LiveCodeBench———
552Llama 3.2 Instruct 1BMetaMeta
2.0
LiveCodeBench60K tokens—✅
553Granite 4.0 350MIBM
2.0
LiveCodeBench———
554Gemma 3 1B InstructGoogleGoogle
2.0
LiveCodeBench———
555LFM2.5-1.2B-InstructLiquid AI
2.0
SciCode———
556Gemma 3 270MGoogleGoogle
0.0
LiveCodeBench———
557Llama 2 Chat 7BMetaMeta
0.0
LiveCodeBench———
558Qwen3.5 0.8B (Reasoning)AlibabaAlibaba
0.0
SciCode———

+ 362 modelos sem benchmark de coding disponível.Ver todos os modelos

Guia Completo: IA para Programação em 2026

O Estado da IA para Código em 2026

A inteligência artificial transformou radicalmente o desenvolvimento de software nos últimos anos. Em 2026, modelos de linguagem (LLMs) são capazes de gerar código funcional em dezenas de linguagens, resolver bugs em projetos reais e até criar aplicações completas a partir de descrições em linguagem natural. O SWE-bench — o benchmark mais rigoroso para coding — avalia modelos em tarefas reais de engenharia de software extraídas de issues do GitHub.

SWE-bench: O Benchmark de Referência

O SWE-bench (Software Engineering Benchmark) é considerado o padrão ouro para avaliar capacidade de coding de LLMs. Diferente de benchmarks acadêmicos como HumanEval (que testa funções isoladas), o SWE-bench apresenta issues reais de repositórios populares como Django, Flask, scikit-learn e requests. O modelo precisa entender o contexto do projeto, localizar os arquivos relevantes e gerar um patch que resolva o bug — simulando o trabalho real de um desenvolvedor.

A versão “Verified” do SWE-bench (SWE-bench Verified) é curada por engenheiros humanos para garantir que cada tarefa tem uma solução clara e verificável. Os scores neste benchmark são particularmente informativos porque correlacionam fortemente com a experiência real de uso para coding.

HumanEval e LiveCodeBench

HumanEval, criado pela OpenAI, testa a capacidade de gerar funções Python a partir de docstrings. É um benchmark mais simples que o SWE-bench, mas útil para avaliar fluência básica em código. LiveCodeBench adiciona uma camada de complexidade ao testar com problemas que são atualizados regularmente, reduzindo o risco de contaminação (quando o modelo já viu as respostas durante o treinamento).

Como Escolher o Melhor Modelo para Código

A escolha do modelo ideal depende do caso de uso específico. Para autocompletar código em tempo real (Cursor, Copilot), velocidade e latência são mais importantes que score máximo — modelos das classes mini e flash costumam oferecer a melhor relação velocidade/qualidade. Para geração de projetos completos ou debug complexo, os modelos frontier da Anthropic, OpenAI e Google tendem a ser mais adequados, apesar do custo maior.

Para equipes que precisam de controle sobre os dados (compliance, segurança), modelos open source como DeepSeek Coder, Code Llama e StarCoder permitem deploy on-premises com performance competitiva. A decisão entre proprietário e open source envolve tradeoffs de custo, latência, privacidade e qualidade.

Ferramentas de Coding com IA

As principais ferramentas de desenvolvimento assistido por IA em 2026 incluem Cursor, GitHub Copilot, Windsurf e fluxos agentic integrados ao editor. Cada ferramenta usa diferentes modelos por baixo, e a qualidade do código gerado depende diretamente da capacidade do LLM utilizado.

Para desenvolvedores brasileiros, um fator importante é a capacidade do modelo de entender comentários, nomes de variáveis e documentação em português — algo que varia significativamente entre modelos e que não é capturado pelos benchmarks tradicionais em inglês.

Tendências para 2026 e Além

As tendências mais relevantes em IA para código incluem: agentes autônomos de engenharia (que resolvem tarefas complexas sem supervisão), geração de testes automatizados, refatoração inteligente, e integração nativa com pipelines de CI/CD. A fronteira está se movendo de “assistente de código” para “engenheiro autônomo”, com modelos cada vez mais capazes de navegar codebases grandes e tomar decisões arquiteturais.

Perguntas Frequentes

Qual é a melhor IA para programar?

Em 2026, os modelos que lideram em benchmarks de código são Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback). No entanto, a melhor escolha depende do caso de uso: autocompletar código, geração de projetos completos, debug ou code review.

ChatGPT ou Claude para código?

Hoje as opções mais fortes em benchmarks de código são Claude Fable 5.1 (Max, Default Fallback), Claude Fable 5.1 (Xhigh, Default Fallback), Claude Fable 5.1 (High, Default Fallback). Claude tende a brilhar em refactors longos; GPT é forte em geração rápida e tool use. O teste final precisa ser feito no seu próprio repositório.

O que é o SWE-bench?

SWE-bench (Software Engineering Benchmark) avalia a capacidade de modelos de resolver issues reais de repositórios open source no GitHub. É considerado o benchmark mais realista para coding, pois testa resolução de bugs em projetos reais, não exercícios acadêmicos.

Quais LLMs gratuitas são boas para código?

Modelos open source como DeepSeek Coder, Qwen Coder e Code Llama oferecem excelente performance em coding sem custo de API. Podem ser rodados localmente via Ollama ou acessados gratuitamente em plataformas como Together AI e Groq.

Cursor ou GitHub Copilot?

Cursor e Copilot são IDEs/extensões que usam LLMs por baixo. Cursor permite escolher o modelo (Claude, GPT, etc.), enquanto Copilot usa modelos da OpenAI. A qualidade do código gerado depende mais do modelo escolhido do que da ferramenta em si.

Explorar Outras Categorias

Todos os ModelosOpen SourceIAs GratuitasImagemÁudioVídeo🇧🇷 PT-BRCompararMetodologiaModelosFerramentas DevRanking para Devs