| 1 | Claude Sonnet 4 | 100.0% | $0.0003 | 1.5s | 100% |
| 2 | GPT-4o Mini (temp=0) | 100.0% | $0.0000 | 7.2s | 100% |
| 3 | Mistral Small Creative | 97.0% | $0.0000 | 349ms | 66% |
| 4 | DeepSeek V3 (2025-03-24) | 97.0% | $0.0000 | 2.3s | 66% |
| 5 | Gemini 3 Flash (Preview, Reasoning) | 99.0% | $0.0029 | 7.9s | 80% |
| 6 | GPT-4o Mini (temp=1) | 98.0% | $0.0000 | 11.5s | 72% |
| 7 | GPT-5.4 Nano | 95.0% | $0.0000 | 720ms | 56% |
| 8 | DeepSeek V3 (2024-12-26) | 93.0% | $0.0000 | 1.3s | 49% |
| 9 | Claude Opus 4 | 95.0% | $0.0017 | 6.1s | 56% |
| 10 | GPT-5.4 Mini | 92.0% | $0.0001 | 650ms | 46% |
| 11 | Gemini 2.5 Flash Lite (Reasoning) | 93.0% | $0.0004 | 4.2s | 49% |
| 12 | Rocinante 12B | 92.0% | $0.0000 | 2.9s | 46% |
| 13 | Hermes 3 70B | 91.0% | $0.0000 | 760ms | 43% |
| 14 | GPT-4.1 | 91.0% | $0.0002 | 1.2s | 43% |
| 15 | Mistral Small 4 (Reasoning) | 92.0% | $0.0004 | 4.4s | 46% |
| 16 | DeepSeek V3.1 | 91.0% | $0.0001 | 3.0s | 43% |
| 17 | Gemma 3 4B | 90.0% | $0.0000 | 256ms | 40% |
| 18 | Gemini 2.5 Flash Lite | 90.0% | $0.0000 | 352ms | 40% |
| 19 | Ministral 3 14B | 90.0% | $0.0000 | 416ms | 40% |
| 20 | Gemma 3 12B | 90.0% | $0.0000 | 443ms | 40% |
| 21 | Gemini 2.5 Flash | 90.0% | $0.0000 | 465ms | 40% |
| 22 | Inception Mercury | 90.0% | $0.0000 | 529ms | 40% |
| 23 | Mistral Small 3.2 24B | 90.0% | $0.0000 | 566ms | 40% |
| 24 | Llama 3.1 70B | 90.0% | $0.0001 | 522ms | 40% |
| 25 | Gemma 3 27B | 90.0% | $0.0000 | 678ms | 40% |
| 26 | Mistral Medium 3.1 | 90.0% | $0.0000 | 634ms | 40% |
| 27 | Gemini 3.1 Flash Lite (Preview) | 90.0% | $0.0000 | 703ms | 40% |
| 28 | Mistral Large 2 | 90.0% | $0.0002 | 437ms | 40% |
| 29 | Qwen 2.5 72B | 90.0% | $0.0000 | 706ms | 40% |
| 30 | Inception Mercury 2 | 90.0% | $0.0002 | 481ms | 40% |
| 31 | Llama 3.1 Nemotron 70B | 90.0% | $0.0000 | 814ms | 40% |
| 32 | Gemini 3 Flash (Preview) | 90.0% | $0.0000 | 826ms | 40% |
| 33 | GPT-4.1 Nano | 90.0% | $0.0000 | 1.0s | 40% |
| 34 | Mistral Large 3 | 90.0% | $0.0000 | 989ms | 40% |
| 35 | Arcee AI: Trinity Large (Preview) | 90.0% | $0.0000 | 1.1s | 40% |
| 36 | GPT-5.4 | 90.0% | $0.0003 | 660ms | 40% |
| 37 | Gemini 2.5 Flash (Reasoning) | 94.0% | $0.0035 | 7.0s | 53% |
| 38 | Claude Haiku 4.5 | 90.0% | $0.0001 | 1.1s | 40% |
| 39 | GPT-4o, Aug. 6th (temp=1) | 90.0% | $0.0002 | 955ms | 40% |
| 40 | GPT-4o, Aug. 6th (temp=0) | 90.0% | $0.0002 | 970ms | 40% |
| 41 | GPT-4.1 Mini | 90.0% | $0.0000 | 1.4s | 40% |
| 42 | DeepSeek-V2 Chat | 90.0% | $0.0000 | 1.5s | 40% |
| 43 | Claude Sonnet 4.6 | 90.0% | $0.0004 | 1.1s | 40% |
| 44 | GPT-5.4 Nano (Reasoning, Low) | 90.0% | $0.0001 | 1.9s | 40% |
| 45 | ByteDance Seed 1.6 Flash | 90.0% | $0.0001 | 2.2s | 40% |
| 46 | Claude 3.7 Sonnet | 90.0% | $0.0004 | 1.7s | 40% |
| 47 | Claude Sonnet 4.5 | 90.0% | $0.0003 | 1.9s | 40% |
| 48 | GPT-5.4 Mini (Reasoning, Low) | 90.0% | $0.0002 | 2.1s | 40% |
| 49 | Z.AI GLM 4.5 | 90.0% | $0.0001 | 2.6s | 40% |
| 50 | GPT-5.4 Nano (Reasoning) | 90.0% | $0.0001 | 2.8s | 40% |
| 51 | WizardLM 2 8x22b | 91.0% | $0.0001 | 6.2s | 43% |
| 52 | Mistral Small 4 | 89.0% | $0.0000 | 527ms | 37% |
| 53 | Claude Opus 4.6 | 90.0% | $0.0006 | 2.4s | 40% |
| 54 | Hermes 3 405B | 91.0% | $0.0000 | 6.4s | 43% |
| 55 | ByteDance Seed 2.0 Lite | 93.0% | $0.0009 | 11.1s | 49% |
| 56 | GPT-5.2 | 91.0% | $0.0016 | 3.7s | 43% |
| 57 | Claude Opus 4.5 | 90.0% | $0.0008 | 2.2s | 40% |
| 58 | GPT-5.4 Mini (Reasoning) | 90.0% | $0.0005 | 2.9s | 40% |
| 59 | GPT-5.4 (Reasoning, Low) | 90.0% | $0.0009 | 2.3s | 40% |
| 60 | Claude 3.5 Haiku | 89.0% | $0.0001 | 1.0s | 37% |
| 61 | Qwen3 235B A22B Instruct 2507 | 89.0% | $0.0000 | 1.4s | 37% |
| 62 | GPT-5 Mini | 91.0% | $0.0008 | 5.7s | 43% |
| 63 | Grok 4 Fast | 90.0% | $0.0003 | 3.9s | 40% |
| 64 | Grok 4.20 (Beta) | 89.0% | $0.0002 | 1.6s | 37% |
| 65 | Mistral Large | 91.0% | $0.0011 | 6.0s | 43% |
| 66 | Claude 3 Haiku | 90.0% | $0.0000 | 5.4s | 40% |
| 67 | Writer: Palmyra X5 | 90.0% | $0.0002 | 5.3s | 40% |
| 68 | Qwen 3.5 Plus (2026-02-15) | 89.0% | $0.0000 | 2.9s | 37% |
| 69 | Claude 3.5 Sonnet | 90.0% | $0.0003 | 5.4s | 40% |
| 70 | Llama 3.1 8B | 88.0% | $0.0000 | 427ms | 35% |
| 71 | GPT-5.1 | 90.0% | $0.0013 | 4.3s | 40% |
| 72 | GPT-4o, May 13th (temp=1) | 90.0% | $0.0004 | 6.0s | 40% |
| 73 | GPT-4o, May 13th (temp=0) | 90.0% | $0.0004 | 6.2s | 40% |
| 74 | GPT-5.4 (Reasoning) | 90.0% | $0.0017 | 3.8s | 40% |
| 75 | Arcee AI: Trinity Mini | 89.0% | $0.0001 | 4.6s | 37% |
| 76 | Grok 4.1 Fast | 90.0% | $0.0004 | 7.0s | 40% |
| 77 | Nemotron 3 Super | 90.0% | $0.0000 | 7.9s | 40% |
| 78 | Stealth: Healer Alpha | 90.0% | $0.0000 | 7.9s | 40% |
| 79 | GPT-5 Nano | 90.0% | $0.0002 | 7.7s | 40% |
| 80 | Claude Sonnet 4.6 (Reasoning) | 90.0% | $0.0025 | 3.4s | 40% |
| 81 | Mistral NeMO | 88.0% | $0.0000 | 3.3s | 35% |
| 82 | Z.AI GLM 5 Turbo | 90.0% | $0.0017 | 6.1s | 40% |
| 83 | Nemotron 3 Nano | 90.0% | $0.0001 | 9.5s | 40% |
| 84 | ByteDance Seed 1.6 | 90.0% | $0.0007 | 8.5s | 40% |
| 85 | Qwen 3.5 Flash | 90.0% | $0.0005 | 8.9s | 40% |
| 86 | Qwen 3 32B | 90.0% | $0.0002 | 9.9s | 40% |
| 87 | Claude Opus 4.6 (Reasoning) | 90.0% | $0.0041 | 4.7s | 40% |
| 88 | Z.AI GLM 4.7 Flash | 90.0% | $0.0003 | 12.5s | 40% |
| 89 | Qwen 3.5 35B | 90.0% | $0.0025 | 8.4s | 40% |
| 90 | DeepSeek V3.2 | 89.0% | $0.0002 | 10.5s | 37% |
| 91 | Gemini 2.5 Pro | 94.0% | $0.0093 | 9.0s | 53% |
| 92 | Grok 4.20 (Beta, Reasoning) | 90.0% | $0.0050 | 5.0s | 40% |
| 93 | MiniMax M2.5 | 91.0% | $0.0009 | 17.2s | 43% |
| 94 | GPT-5 | 90.0% | $0.0041 | 8.5s | 40% |
| 95 | MiniMax M2.7 | 91.0% | $0.0014 | 16.8s | 43% |
| 96 | Stealth: Hunter Alpha | 91.0% | $0.0000 | 19.9s | 43% |
| 97 | Stealth: Aurora Alpha | 90.0% | — | 1.5s | 40% |
| 98 | Z.AI GLM 4.6 | 95.0% | $0.0018 | 31.4s | 56% |
| 99 | Qwen 3.5 27B | 90.0% | $0.0030 | 13.8s | 40% |
| 100 | o4 Mini | 91.0% | $0.0020 | 19.5s | 43% |
| 101 | Ministral 8B | 82.0% | $0.0000 | 305ms | 23% |
| 102 | o4 Mini High | 90.0% | $0.0047 | 12.5s | 40% |
| 103 | Aion 2.0 | 90.0% | $0.0014 | 21.0s | 40% |
| 104 | Cohere Command R+ (Aug. 2024) | 81.0% | $0.0002 | 447ms | 22% |
| 105 | Qwen 3.5 9B | 90.0% | $0.0002 | 24.5s | 40% |
| 106 | Ministral 3 3B | 80.0% | $0.0000 | 360ms | 20% |
| 107 | MoonshotAI: Kimi K2.5 | 92.0% | $0.0032 | 26.0s | 46% |
| 108 | LFM2 24B | 80.0% | $0.0000 | 1.3s | 20% |
| 109 | Ministral 3B | 79.0% | $0.0000 | 297ms | 19% |
| 110 | ByteDance Seed 2.0 Mini | 90.0% | $0.0005 | 31.5s | 40% |
| 111 | Qwen 3.5 122B | 90.0% | $0.0085 | 21.6s | 40% |
| 112 | Gemini 3 Pro (Preview) | 90.0% | $0.015 | 12.9s | 40% |
| 113 | Qwen 3.5 397B A17B | 90.0% | $0.0048 | 34.5s | 40% |
| 114 | Z.AI GLM 4.7 | 90.0% | $0.0021 | 42.9s | 40% |
| 115 | Ministral 3 8B | 70.0% | $0.0000 | 367ms | 8% |
| 116 | Z.AI GLM 5 | 92.0% | $0.0046 | 46.7s | 46% |
| 117 | Gemini 3.1 Pro (Preview) | 92.0% | $0.024 | 22.8s | 46% |
| 118 | Grok 4 | 90.0% | $0.019 | 30.3s | 40% |