Ranking de Modelos de IA · Blog do Bufano

OpenRouter

Uso real — quais modelos os desenvolvedores mais utilizam, medido por tokens processados.

Ver fonte → 01 de jul. de 26, 03:18

1 Anthropic: Claude Sonnet 5 Anthropic vision —

2 Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) Google visionfastest —

3 Sakana: Fugu Ultra Sakana vision —

4 Google: Nano Banana 2 (Gemini 3.1 Flash Image) Google visionfastest —

5 Google: Nano Banana Pro (Gemini 3 Pro Image) Google visionagentic —

6 Cohere: North Mini Code (free) Cohere codingfastest —

7 Z.ai: GLM 5.2 Z.ai —

8 OpenRouter: Fusion OpenRouter —

9 MoonshotAI: Kimi K2.7 Code MoonshotAI visioncoding —

10 Anthropic: Claude Fable Latest Anthropic vision —

11 Anthropic: Claude Fable 5 Anthropic vision —

12 Nex AGI: Nex-N2-Pro Nex AGI vision —

13 NVIDIA: Nemotron 3.5 Content Safety (free) NVIDIA visionreasoning —

14 NVIDIA: Nemotron 3 Ultra (free) NVIDIA reasoning —

15 NVIDIA: Nemotron 3 Ultra NVIDIA reasoning —

16 Qwen: Qwen3.7 Plus Qwen vision —

17 MiniMax: MiniMax M3 MiniMax vision —

18 StepFun: Step 3.7 Flash StepFun visionfastest —

19 Anthropic: Claude Opus 4.8 (Fast) Anthropic visionfastest —

20 Anthropic: Claude Opus 4.8 Anthropic vision —

21 Qwen: Qwen3.7 Max Qwen —

22 xAI: Grok Build 0.1 xAI vision —

23 Google: Gemini 3.5 Flash Google visionfastest —

24 Anthropic: Claude Opus 4.7 (Fast) Anthropic visionfastest —

25 Perceptron: Perceptron Mk1 Perceptron vision —

26 inclusionAI: Ring-2.6-1T inclusionAI —

27 Google: Gemini 3.1 Flash Lite Google visionfastest —

28 OpenAI: GPT Chat Latest OpenAI vision —

29 xAI: Grok 4.3 xAI vision —

30 IBM: Granite 4.1 8B IBM —

31 Mistral: Mistral Medium 3.5 Mistral vision —

32 NVIDIA: Nemotron 3 Nano Omni (free) NVIDIA visionreasoning —

33 Poolside: Laguna XS.2 (free) Poolside —

34 Poolside: Laguna XS.2 Poolside —

35 Poolside: Laguna M.1 (free) Poolside —

36 Poolside: Laguna M.1 Poolside —

37 Anthropic Claude Haiku Latest visionfastest —

38 OpenAI GPT Mini Latest visionfastest —

39 Google Gemini Pro Latest vision —

40 MoonshotAI Kimi Latest vision —

41 Google Gemini Flash Latest visionfastest —

42 Anthropic Claude Sonnet Latest vision —

43 OpenAI GPT Latest vision —

44 Qwen: Qwen3.5 Plus 2026-04-20 Qwen vision —

45 Qwen: Qwen3.6 Flash Qwen visionfastest —

46 Qwen: Qwen3.6 35B A3B Qwen vision —

47 Qwen: Qwen3.6 Max Preview Qwen —

48 Qwen: Qwen3.6 27B Qwen vision —

49 OpenAI: GPT-5.5 Pro OpenAI vision —

50 OpenAI: GPT-5.5 OpenAI vision —

51 DeepSeek: DeepSeek V4 Pro DeepSeek —

52 DeepSeek: DeepSeek V4 Flash DeepSeek fastest —

53 inclusionAI: Ling-2.6-1T inclusionAI —

54 Tencent: Hy3 preview Tencent —

55 Xiaomi: MiMo-V2.5-Pro Xiaomi —

56 Xiaomi: MiMo-V2.5 Xiaomi vision —

57 OpenAI: GPT-5.4 Image 2 OpenAI vision —

58 inclusionAI: Ling-2.6-flash inclusionAI fastest —

59 Anthropic: Claude Opus Latest Anthropic vision —

60 Pareto Code Router coding —

61 MoonshotAI: Kimi K2.6 MoonshotAI vision —

62 Anthropic: Claude Opus 4.7 Anthropic vision —

63 Z.ai: GLM 5.1 Z.ai —

64 Google: Gemma 4 26B A4B (free) Google vision —

65 Google: Gemma 4 26B A4B Google vision —

66 Google: Gemma 4 31B (free) Google vision —

67 Google: Gemma 4 31B Google vision —

68 Qwen: Qwen3.6 Plus Qwen vision —

69 Z.ai: GLM 5V Turbo Z.ai visionfastest —

70 Arcee AI: Trinity Large Thinking Arcee AI reasoning —

71 xAI: Grok 4.20 Multi-Agent xAI visionagentic —

72 xAI: Grok 4.20 xAI vision —

73 Google: Lyria 3 Pro Preview Google vision —

74 Google: Lyria 3 Clip Preview Google vision —

75 Kwaipilot: KAT-Coder-Pro V2 Kwaipilot coding —

76 Reka Edge vision —

77 MiniMax: MiniMax M2.7 MiniMax —

78 OpenAI: GPT-5.4 Nano OpenAI vision —

79 OpenAI: GPT-5.4 Mini OpenAI visionfastest —

80 Mistral: Mistral Small 4 Mistral vision —

81 Z.ai: GLM 5 Turbo Z.ai fastest —

82 NVIDIA: Nemotron 3 Super (free) NVIDIA reasoning —

83 NVIDIA: Nemotron 3 Super NVIDIA reasoning —

84 ByteDance Seed: Seed-2.0-Lite ByteDance Seed visionfastest —

85 Qwen: Qwen3.5-9B Qwen vision —

86 OpenAI: GPT-5.4 Pro OpenAI vision —

87 OpenAI: GPT-5.4 OpenAI vision —

88 Inception: Mercury 2 Inception —

89 OpenAI: GPT-5.3 Chat OpenAI vision —

90 Google: Gemini 3.1 Flash Lite Preview Google visionfastest —

91 ByteDance Seed: Seed-2.0-Mini ByteDance Seed visionfastest —

92 Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview) Google visionfastest —

93 Qwen: Qwen3.5-35B-A3B Qwen vision —

94 Qwen: Qwen3.5-27B Qwen vision —

95 Qwen: Qwen3.5-122B-A10B Qwen vision —

96 Qwen: Qwen3.5-Flash Qwen visionfastest —

97 LiquidAI: LFM2-24B-A2B LiquidAI —

98 Google: Gemini 3.1 Pro Preview Custom Tools Google visionagentic —

99 OpenAI: GPT-5.3-Codex OpenAI vision —

100 AionLabs: Aion-2.0 AionLabs —

Chatbot Arena

Elo crowd-sourced de milhões de comparações cegas head-to-head, votadas por humanos.

Ver fonte → 01 de jul. de 26, 03:18

1 Claude Fable 5 (High) Anthropic · Proprietary most battle-tested — 16,082

2 Claude Opus 4.8 (Thinking) Anthropic · Proprietary most battle-tested — 30,511

3 GPT 5.5 (xHigh) OpenAI · Proprietary most battle-tested — 24,393

4 Claude Opus 4.7 Anthropic · Proprietary most battle-tested — 31,725

5 Claude Opus 4.7 (Thinking) Anthropic · Proprietary most battle-tested — 31,304

6 GPT 5.5 (High) OpenAI · Proprietary most battle-tested — 49,559

7 GLM 5.2 (Max) Z.ai · MIT · SiliconFlow most battle-tested — 21,946

8 GPT 5.4 (High) OpenAI · Proprietary most battle-tested — 49,486

9 Claude Opus 4.6 Anthropic · Proprietary most battle-tested — 31,155

10 GPT 5.5 OpenAI · Proprietary most battle-tested — 49,883

11 Claude Opus 4.8 Anthropic · Proprietary most battle-tested — 28,284

12 Claude Sonnet 4.6 Anthropic · Proprietary most battle-tested — 31,694

13 GLM 5.1 Z.ai · MIT · SiliconFlow most battle-tested — 40,253

14 Kimi K2.7 Code Moonshot · Modified MIT · Fireworks most battle-tested — 26,189

15 Gemini 3.1 Pro Preview Google · Proprietary most battle-tested — 49,868

16 Gemini 3.5 Flash Google · Proprietary most battle-tested — 44,029

17 DeepSeek V4 Flash DeepSeek · MIT · SiliconFlow most battle-tested — 38,964

18 Kimi K2.6 Moonshot · Modified MIT · Fireworks most battle-tested — 46,299

19 Minimax M3 MiniMax · Proprietary · Fireworks most battle-tested — 31,887

20 DeepSeek V4 Pro DeepSeek · MIT · SiliconFlow most battle-tested — 28,972

21 Qwen 3.6 Plus Alibaba · Proprietary · Fireworks most battle-tested — 41,967

22 Grok 4.3 (High) xAI · Proprietary most battle-tested — 29,641

23 Grok Build 0.1 xAI · Proprietary most battle-tested — 40,828

24 Gemini 3 Flash Google · Proprietary most battle-tested — 49,944

25 Minimax M2.7 MiniMax · Modified MIT · Fireworks most battle-tested — 43,484

26 Nemotron 3 Ultra Nvidia · OpenMDW-1.1 most battle-tested — 7,048

27 Gemma 4 31B Google · Apache 2.0 most battle-tested — 39,388

28 Grok 4.3 xAI · Proprietary most battle-tested — 49,309

Artificial Analysis

Índice independente de inteligência, código e capacidade agêntica, com velocidade e preço.

Ver fonte → 01 de jul. de 26, 03:19

1 Claude Fable 5 (with fallback) 1

2 Claude Opus 4.8 (max) 1

3 GPT-5.5 (xhigh) 922

4 Claude Opus 4.7 (max) 1

5 Claude Sonnet 5 (max) 1

6 GPT-5.5 (high) 922

7 GLM-5.2 (max) 1

8 GPT-5.5 (medium) 922

9 Gemini 3.5 Flash 1

10 Claude Sonnet 4.6 (max) 1

11 Gemini 3.1 Pro Preview 1

12 Qwen3.7 Max 1

13 Gemini 3.5 Flash (medium) 1

14 MiniMax-M3 1

15 DeepSeek V4 Pro (Max) 1

16 GPT-5.3 Codex (xhigh) 400

17 GPT-5.5 (low) 922

18 Muse Spark 262

19 Kimi K2.6 256

20 Claude Opus 4.7 (Non-reasoning, high) 1

21 MiMo-V2.5-Pro 1

22 Kimi K2.7 Code 256

23 Nex-N2-Pro 262

24 DeepSeek V4 Pro (High) 1

25 DeepSeek V4 Flash (Max) 1

26 GLM-5.1 200

27 MiMo-V2.5 1

28 GPT-5.4 mini (xhigh) 400

29 Grok Build 0.1 0616 256

30 Qwen3.6 Plus 1

31 Qwen3.7 Plus 1

32 GPT-5.4 nano (xhigh) 400

33 MiniMax-M2.7 205

34 GLM-5-Turbo 200

35 Nemotron 3 Ultra 262

36 Grok 4.3 (high) 1

37 DeepSeek V4 Flash (High) 1

38 Qwen3.6 27B 262

39 MiMo-V2-Omni-0327 256

40 Grok 4.3 (medium) 1

41 Claude Sonnet 4.6 (Non-reasoning) 1

42 Grok 4.3 (low) 1

43 GPT-5.5 (Non-reasoning) 922

44 GLM-5.1 200

45 MiMo-V2-Omni 256

46 Gemini 3.5 Flash (minimal) 1

47 Kimi K2.6 256

48 GLM 5V Turbo 200

49 Claude Sonnet 4.6 (Non-reasoning, Low Effort) 1

50 Qwen3.5 397B A17B 262

51 Hy3-preview 256

52 MiMo-V2-Flash (Feb 2026) 256

53 Qwen3.5 122B A10B 262

54 Qwen3.5 397B A17B 262

55 Qwen3.6 35B A3B 262

56 DeepSeek V4 Pro 1

57 Qwen3.5 Omni Plus 256

58 Ring-2.6-1T 262

59 o3 200

60 GPT-5.4 nano 400

61 Mistral Medium 3.5 256

62 GPT-5.4 mini (medium) 400

63 Step 3.7 Flash 262

64 Claude 4.5 Haiku 200

65 Gemma 4 31B 256

66 Command A+ 192

67 Qwen3.6 27B 262

68 GPT-5.5 Instant (June 2026) 400

69 DeepSeek V4 Flash 1

70 JT-35B-Flash 256

71 Qwen3.5 122B A10B 262

72 MiMo-V2.5-Pro 1

73 Hy3-preview 256

74 Ling-2.6-1T 262

75 Step 3.5 Flash 2603 256

76 Doubao Seed Code 256

77 Gemini 2.5 Pro 1

78 Gemma 4 26B A4B 256

79 NVIDIA Nemotron 3 Super 1

80 Mercury 2 128

81 Gemini 3.1 Flash-Lite 1

82 Gemma 4 31B 256

83 Grok 4.3 (Non-reasoning) 1

84 K-EXAONE 256

85 MiMo-V2-Flash 256

86 Trinity Large Thinking 512

87 Qwen3.6 35B A3B 262

88 gpt-oss-120b (high) 131

89 Claude 4.5 Haiku 200

90 Qwen3.5 35B A3B 262

91 EXAONE 4.5 33B 262

92 HyperNova 60B 2605 131

93 Gemma 4 12B 256

94 ERNIE 5.0 Thinking Preview 128

95 Nova 2.0 Pro Preview (medium) 256

96 Qwen3.5 9B 262

97 Nemotron Cascade 2 30B A3B 1

98 Qwen3 Coder Next 256

99 Nova 2.0 Omni (medium) 1

100 Mistral Small 4 256

LLM Stats

Agregador de benchmarks compostos (GPQA, SWE-bench, HLE) com preço e throughput em tempo real.

Ver fonte → 19 de jun. de 26, 20:05

1 Claude Mythos PreviewUNRELEASEDAnthropic 65

2 GPT-5.2 ProOpenAI 63

3 Claude Opus 4.8Anthropic 61

4 GLM-5.2NEWZhipu AI 59

5 Claude Fable 5UNRELEASEDAnthropic 59

6 GPT-5.4OpenAI 57

7 GPT-5.5OpenAI 56

8 GPT-5.1OpenAI 56

9 Seed 2.0 ProByteDance 56

10 GPT-5.1 InstantOpenAI 55

11 Qwen3.7 MaxAlibaba Cloud / Qwen Team 54

12 Claude Opus 4.7Anthropic 54

13 GPT-5 MediumOpenAI 54

14 Gemini 3.5 FlashGoogle 53

15 Gemini 3.1 ProGoogle 53

HuggingFace · LMArena

O leaderboard do LMArena hospedado no HuggingFace, rankeado por volume de votos da comunidade.

Ver fonte →

Dados não disponíveis ainda.
Aguarde a próxima atualização.

Ranking de Modelos de IA.

OpenRouter

Chatbot Arena

Artificial Analysis

LLM Stats

HuggingFace · LMArena