腾讯混元 Hy4 在 Artificial Analysis Cyber Index 排名第三、CWE-Bench-AA 登顶
腾讯混元 Hy4 在 Artificial Analysis Cyber Index 以 58% 得分排名第三，并以 76% 登顶 CWE-Bench-AA，领先 DeepSeek V4.1 Flash 1017 (max) 6 分。
Tencent Hy· @TencentHunyuan · X·2026-11-13 16:39· 1 小时前AI 评分57
AI 导读
腾讯混元转发 Artificial Analysis 评测，Hy4 以 58% 得分在 Artificial Analysis Cyber Index 排名第三，并以 76% 登顶 CWE-Bench-AA，领先 DeepSeek V4.1 Flash 1017 (max) 6 分。Hy4 单任务成本 $0.92，处于能力成本 Pareto 前沿，比 Grok 4.7 (xhigh) 高 2 分且成本仅 1/11；每任务输出 64k token，比 DeepSeek V4.1 Flash 1017 (max) 少 43%。API 定价为 $0.834/$2.501 per 1M input/output tokens。
正文 · 原文
#3 on the Artificial Analysis Cyber Index and #1 on CWE-Bench-AA 🎉🎉🎉
在 X 查看被引用的帖子Tencent's Hy4 ranks #3 on the Artificial Analysis Cyber Index and #1 on CWE-Bench-AA
@TencentHunyuan's Hy4 scores 58% on the Artificial Analysis Cyber Index, 1 point ahead of DeepSeek V4.1 Flash 1017 (max). It trails Kimi K3.1 Pro (64%) and the trusted-access GPT-6 Sol (Daybreak Blue, max) (69%). API pricing is $0.834/$2.501 per 1M input/output tokens.
Key results:
- #1 on CWE-Bench-AA (76%). Hy4 leads DeepSeek V4.1 Flash 1017 (max) by 6 points on our implementation of @CollinearAI's CWE-bench v1, which tests agents' ability to find and fix real software vulnerabilities
- Strong capability for the cost. At $0.92 per task, Hy4 sits on the Capability vs. Cost Pareto frontier. It scores 2 points above Grok 4.7 (xhigh) at 1/11th the cost and 1 point above DeepSeek V4.1 Flash 1017 (max) at 2x the cost
- Fewer output tokens than peers. Hy4 uses 64k output tokens per task: 43% fewer than DeepSeek V4.1 Flash 1017 (max) and 37% fewer than Grok 4.7 (xhigh), but 49% more than Kimi K3.1 Pro and 56% more than GPT-6 Sol (Daybreak Blue, max)
来源：Tencent Hy · x.com
来源：RadarAI · aihot.news