Name: 智能体模型厂商排行榜（2026-06-12）- 最佳 Agent 模型对比 - 厂商聚合排名
Creator: LM Arena
License: https://creativecommons.org/licenses/by/4.0/
Keywords: AI 智能体模型排行榜,智能体 AI 排行榜,最佳 Agent 模型对比,AI 模型排行,智能体评测,净提升

排名Rank ⇕	厂商Lab ⇕
› 🥇	AnthropicAnthropic Claude Fable 5 (High) · Proprietary	13.68% ±1.59%

综合排名Overall Rank — 名次区间Rank Spread 1 ↔ 2 确认成功Confirmed Success ▲17.21%±2.59% 赞扬/抱怨Praise vs Complaint ▲27.74%±5.97% 可控性Steerability ▲11.27%±3.24% Bash 恢复Bash Recovery ▲10.16%±1.40% 工具幻觉Tool Hallucination ▼2.00%±0.24% 会话数Sessions 16,167
› 🥈	OpenAIOpenAI GPT 5.5 (xHigh) · Proprietary	11.03% ±2.52%

综合排名Overall Rank — 名次区间Rank Spread 1 ↔ 7 确认成功Confirmed Success ▲5.69%±4.32% 赞扬/抱怨Praise vs Complaint ▲27.74%±9.54% 可控性Steerability ▲5.05%±5.05% Bash 恢复Bash Recovery ▲14.67%±1.76% 工具幻觉Tool Hallucination ▼2.00%±0.24% 会话数Sessions 5,861
› 🥉	智谱 ZAIZ.ai GLM 5.1 · MIT · SiliconFlow	2.88% ±1.21%

综合排名Overall Rank — 名次区间Rank Spread 10 ↔ 12 确认成功Confirmed Success ▲5.16%±2.46% 赞扬/抱怨Praise vs Complaint ▲1.22%±4.21% 可控性Steerability ▲1.09%±2.46% Bash 恢复Bash Recovery ▲4.94%±1.38% 工具幻觉Tool Hallucination ▼2.00%±0.24% 会话数Sessions 27,536
› 4	DeepSeekDeepSeek DeepSeek V4 Pro · MIT · SiliconFlow	0.22% ±1.18%

综合排名Overall Rank — 名次区间Rank Spread 13 ↔ 17 确认成功Confirmed Success ▲2.42%±2.50% 赞扬/抱怨Praise vs Complaint ▲1.35%±4.12% 可控性Steerability ▲2.74%±2.46% Bash 恢复Bash Recovery ▲2.15%±1.10% 工具幻觉Tool Hallucination ▼0.61%±0.37% 会话数Sessions 26,839
› 5	GoogleGoogle Gemini 3.5 Flash · Proprietary	0.06% ±1.06%

综合排名Overall Rank — 名次区间Rank Spread 13 ↔ 17 确认成功Confirmed Success ▲2.85%±2.38% 赞扬/抱怨Praise vs Complaint ▲1.65%±3.55% 可控性Steerability ▲0.24%±2.02% Bash 恢复Bash Recovery ▲3.08%±1.27% 工具幻觉Tool Hallucination ▼1.94%±0.24% 会话数Sessions 25,513
› 6	月之暗面Moonshot Kimi K2.6 · Modified MIT · Fireworks	0.63% ±1.09%

综合排名Overall Rank — 名次区间Rank Spread 13 ↔ 18 确认成功Confirmed Success ▲0.62%±2.34% 赞扬/抱怨Praise vs Complaint ▲4.08%±3.66% 可控性Steerability ▲3.14%±2.25% Bash 恢复Bash Recovery ▲1.45%±1.49% 工具幻觉Tool Hallucination ▼2.00%±0.24% 会话数Sessions 29,002
› 7	阿里巴巴Alibaba Qwen 3.6 Plus · Proprietary · Fireworks	4.52% ±1.21%

综合排名Overall Rank — 名次区间Rank Spread 18 ↔ 20 确认成功Confirmed Success ▲1.74%±2.51% 赞扬/抱怨Praise vs Complaint ▲8.04%±4.02% 可控性Steerability ▲9.57%±2.61% Bash 恢复Bash Recovery ▲1.79%±1.75% 工具幻觉Tool Hallucination ▼1.47%±0.59% 会话数Sessions 27,118
› 8	xAIxAI Grok Build 0.1 · Proprietary	6.28% ±1.11%

综合排名Overall Rank — 名次区间Rank Spread 18 ↔ 22 确认成功Confirmed Success ▲6.19%±2.62% 赞扬/抱怨Praise vs Complaint ▲12.45%±3.51% 可控性Steerability ▲9.84%±2.37% Bash 恢复Bash Recovery ▲1.93%±1.60% 工具幻觉Tool Hallucination ▼4.85%±0.69% 会话数Sessions 22,542
› 9	NVIDIANvidia Nemotron 3 Ultra · OpenMDW-1.1	6.81% ±5.22%

综合排名Overall Rank — 名次区间Rank Spread 15 ↔ 24 确认成功Confirmed Success ▲3.00%±8.86% 赞扬/抱怨Praise vs Complaint ▲2.58%±19.26% 可控性Steerability ▲23.87%±10.36% Bash 恢复Bash Recovery ▲11.75%±9.55% 工具幻觉Tool Hallucination ▼2.00%±0.24% 会话数Sessions 2,711
› 10	MiniMaxMiniMax Minimax M2.7 · Modified MIT · Fireworks	7.43% ±1.10%

综合排名Overall Rank — 名次区间Rank Spread 19 ↔ 23 确认成功Confirmed Success ▲10.72%±2.65% 赞扬/抱怨Praise vs Complaint ▲16.78%±3.36% 可控性Steerability ▲9.09%±2.16% Bash 恢复Bash Recovery ▲2.51%±1.77% 工具幻觉Tool Hallucination ▼1.96%±0.25% 会话数Sessions 27,792

没有找到相关模型No matching models found

AI 智能体模型排行榜Agent Models Leaderboard 🏆 综合榜单