HyperAI
Command Palette
Search for a command to run...
FrontierCode-AA 2.0
Cognition FrontierCode 2.0 的实现，这是一个私有智能体编程基准，包含真实开源仓库中的 132 个任务，以维护者是否会合并智能体的改动作为评分标准。
- 类别
- 编程
- 智能指数权重
- 15%
- 模型
- 107 当前，123 总共
- 中位数
- 20.2%
- 范围
- 0.0%–49.1%
| 排名 | 比较 | 模型 | 评分 | 成本 / 任务 | 时间 / 任务 | 输出 token / 任务 |
|---|---|---|---|---|---|---|
| 51 |  MiMo-V2.6-Flash Xiaomi | 21.6% | $0.11 | 56m 37s | 119K | |
| 52 |  GLM-5.3 (Max) Z AI | 21.6% | $6.10 | 25m 42s | 106K | |
| 53 |  MiniMax-M3.1-Flash MiniMax | 20.8% | $0.47 | 13m 29s | 129K | |
| 54 |  GPT-6.1 Sol (Low) OpenAI | 20.2% | $0.20 | 1m 50s | 6.23K | |
| 55 |  Gemini 3.8 Flash (High) Google | 18.7% | $14.90 | 17m 7s | 118K | |
| 56 | GPT-6 Luna (High) OpenAI | 18.6% | $0.084 | 5m 6s | 29.6K | |
| 57 | GPT-6 Luna (Xhigh) OpenAI | 18.4% | $0.11 | 6m 12s | 38.3K | |
| 58 | GLM-5.3-Flash Z AI | 17.6% | $0.45 | 26m 13s | 77.9K | |
| 59 |  DeepSeek V4 Pro 0813 (Max) DeepSeek | 17.1% | $1.83 | 25m 36s | 106K | |
| 60 |  Ling 3.1 Flash 1019 InclusionAI | 17.0% | $2.36 | — | 148K | |
| 61 | DeepSeek V4.1 Flash 1017 (High) DeepSeek | 16.8% | $0.35 | 11m 57s | 102K | |
| 62 |  Agnes 2.5 Pro Theta Sapiens AI | 16.6% | $0.68 | 33m 55s | 393K | |
| 63 | 16.5% | $11.61 | 15m 4s | 101K | ||
| 64 |  Claude Haiku 5.5 (Medium) Anthropic | 16.1% | $0.076 | 3m 29s | 47.1K | |
| 65 | 15.8% | $0.63 | 2m 10s | 14.1K | ||
| 66 |  Mistral Large 4 (High) Mistral | 15.0% | $1.09 | 6m 31s | 45.8K | |
| 67 |  Qwen3.8 27B (Xhigh) Alibaba | 14.9% | $4.48 | 39m 48s | 110K | |
| 68 | Mistral Large 4 (Medium) Mistral | 13.3% | $1.02 | 6m 48s | 46K | |
| 69 | GPT-6 Luna (Medium) OpenAI | 12.8% | $0.053 | 2m 53s | 17.3K | |
| 70 | GLM-5.3 (Low) Z AI | 12.7% | $1.51 | 9m 11s | 32.5K | |
| 71 | Qwen3.8 27B (Low) Alibaba | 12.4% | $2.90 | 30m 5s | 90.5K | |
| 72 |  Step 5 Preview StepFun | 12.0% | $2.11 | 9m 21s | 80.9K | |
| 73 |  K2 Horizon 375B A23B Institute of Foundation Models | 11.8% | $0 | 7m 15s | 81.8K | |
| 74 | Qwen3.8 27B (Medium) Alibaba | 11.5% | $3.69 | 38m 39s | 104K | |
| 75 | 11.0% | $0.32 | 1m 36s | 13.3K | ||
| 76 |  Nanbeige4.3-33B Nanbeige | 10.7% | — | — | 168K | |
| 77 | 10.3% | $1.32 | 18m 21s | 73.1K | ||
| 78 | Gemini 3.8 Flash (Low) Google | 8.1% | — | — | — | |
| 79 | Claude Haiku 5.5 (Low) Anthropic | 8.0% | $0.039 | 1m 58s | 25.9K | |
| 80 | 7.6% | $3.01 | 5m 44s | 59.5K | ||
| 81 | 7.1% | $0.25 | 1m 15s | 10.5K | ||
| 82 |  Apodex 1.1 Apodex | 6.9% | $0.99 | 14m 56s | 127K | |
| 83 |  Nemotron 3.5 Super VL NVIDIA | 6.1% | $1.32 | 6m 1s | 93.6K | |
| 84 | 6.1% | $0.87 | 13m 2s | 98.1K | ||
| 85 | Gemini 3.5 Flash-Lite Google | 6.0% | $2.46 | 3m 14s | 60.1K | |
| 86 |  Beam 1.0 Reflection | 5.5% | $0.30 | 1m 47s | 30.8K | |
| 87 |  Inkling Small Thinking Machines | 4.8% | $0.47 | 9m 0s | 38.7K | |
| 88 | K2 Horizon MoVA 36B A4B Institute of Foundation Models | 4.5% | — | — | — | |
| 89 | Inkling (Xhigh) Thinking Machines | 3.9% | $3.27 | 10m 11s | 59.5K | |
| 90 | Qwen3.8 27B (Non-reasoning) Alibaba | 3.7% | $5.52 | 20m 44s | 50.6K | |
| 91 | GPT-6 Luna (Low) OpenAI | 2.2% | $0.0073 | 29.8s | 2.58K | |
| 92 | Mistral Medium 3.5 Mistral | 1.8% | $1.94 | 3m 52s | 32.5K | |
| 93 | 1.6% | $0.13 | 2m 26s | 13.2K | ||
| 94 | 1.5% | $0.47 | 8m 4s | 43.3K | ||
| 95 | K2 Horizon 7B Institute of Foundation Models | 1.2% | — | — | — | |
| 96 | 1.1% | $4.85 | 7m 1s | 79.6K | ||
| 97 | K2 Horizon 3.7B Institute of Foundation Models | 0.9% | — | — | — | |
| 98 |  K-EXAONE 2.0 LG AI Research | 0.8% | — | — | 72.9K | |
| 99 |  North Mini Code Cohere | 0.2% | $0 | 3m 44s | 38K | |
| 100 | Command A+ Cohere | 0.0% | $0 | 4m 26s | 52.9K |
