deepseek-v4.1IMAGE
DeepSeek 1017 tops open models on Agent Arena, ranks 11th overall
On Agent Arena's real agent-task leaderboard, DeepSeek 1017 ranks 11th globally and first among open models, with net improvement rising from 2.56% to 5.13% and user-confirmed completion from 6.60% to 11.54%.
gshota
@gshota666Archived 11 days agoOriginally posted 2026.10.20

gshota @gshota666Originally posted 2026.10.20
🚨 DeepSeek 1017 再传捷报!Agent Arena 开源模型第一,综合排名冲进前 11! Arena 最新真实智能体任务评测: 🔥 DeepSeek 1017:全球第 11 💻 DeepSeek 旧版:全球第 20 📊 综合净改善指标: 旧版:2.56% 新版:5.13% 提高 2.57 个百分点! 📈 用户确认任务完成指标: 旧版:6.60% 新版:11.54% 提高 4.94 个百分点! AI 模型的竞争,正在从刷 Benchmark 转向真正把任务完成。 兄弟们,你们觉得 DeepSeek 现在能成为 Agent 主力模型了吗? 完整榜单:https://arena.ai/leaderboard/agent
x.com/gshota666/status/2112281919350321217Archived 2026.10.20 12:06