Name: 智能体模型厂商排行榜（2026-07-08）- 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	14.10% ±1.56%

综合排名Overall Rank — 名次区间Rank Spread 1 ↔ 1 确认成功Confirmed Success ▲16.88%±2.73% 赞扬/抱怨Praise vs Complaint ▲30.94%±5.70% 可控性Steerability ▲12.56%±2.95% Bash 恢复Bash Recovery ▲8.91%±1.85% 工具幻觉Tool Hallucination ▼1.24%±0.13% 会话数Sessions 16,059
› 🥈	OpenAIOpenAI GPT 5.5 (xHigh) · Proprietary	8.90% ±0.91%

综合排名Overall Rank — 名次区间Rank Spread 2 ↔ 7 确认成功Confirmed Success ▲7.69%±1.73% 赞扬/抱怨Praise vs Complaint ▲13.87%±3.31% 可控性Steerability ▲7.18%±1.84% Bash 恢复Bash Recovery ▲14.54%±0.85% 工具幻觉Tool Hallucination ▼1.24%±0.13% 会话数Sessions 32,695
› 🥉	智谱 ZAIZ.ai GLM 5.2 (Max) · MIT · SiliconFlow	6.54% ±1.17%

综合排名Overall Rank — 名次区间Rank Spread 4 ↔ 12 确认成功Confirmed Success ▲9.46%±2.21% 赞扬/抱怨Praise vs Complaint ▲13.74%±4.23% 可控性Steerability ▲3.44%±2.30% Bash 恢复Bash Recovery ▲4.82%±1.17% 工具幻觉Tool Hallucination ▼1.24%±0.13% 会话数Sessions 29,649
› 4	阿里巴巴Alibaba Qwen3.7 Max · Proprietary	0.36% ±2.01%

综合排名Overall Rank — 名次区间Rank Spread 13 ↔ 22 确认成功Confirmed Success ▲1.43%±4.14% 赞扬/抱怨Praise vs Complaint ▲0.52%±6.44% 可控性Steerability ▲5.20%±4.97% Bash 恢复Bash Recovery ▲7.91%±2.30% 工具幻觉Tool Hallucination ▼1.02%±0.24% 会话数Sessions 6,271
› 5	月之暗面Moonshot Kimi K2.7 Code · Modified MIT	0.13% ±2.60%

综合排名Overall Rank — 名次区间Rank Spread 13 ↔ 23 确认成功Confirmed Success ▲7.70%±4.79% 赞扬/抱怨Praise vs Complaint ▲0.63%±8.70% 可控性Steerability ▲6.82%±6.57% Bash 恢复Bash Recovery ▲0.86%±2.95% 工具幻觉Tool Hallucination ▼1.24%±0.13% 会话数Sessions 4,870
› 6	GoogleGoogle Gemini 3.1 Pro Preview · Proprietary	0.17% ±0.70%

综合排名Overall Rank — 名次区间Rank Spread 15 ↔ 21 确认成功Confirmed Success ▲0.72%±1.54% 赞扬/抱怨Praise vs Complaint ▲0.72%±2.24% 可控性Steerability ▲3.47%±1.28% Bash 恢复Bash Recovery ▲6.98%±1.22% 工具幻觉Tool Hallucination ▼1.20%±0.14% 会话数Sessions 58,057
› 7	DeepSeekDeepSeek DeepSeek V4 Pro · MIT	0.83% ±1.86%

综合排名Overall Rank — 名次区间Rank Spread 14 ↔ 23 确认成功Confirmed Success ▲2.34%±4.09% 赞扬/抱怨Praise vs Complaint ▲3.13%±5.85% 可控性Steerability ▲4.61%±4.73% Bash 恢复Bash Recovery ▲4.94%±1.34% 工具幻觉Tool Hallucination ▼0.99%±0.22% 会话数Sessions 6,680
› 8	MiniMaxMiniMax Minimax M3 · MiniMax Community License	3.08% ±1.78%

综合排名Overall Rank — 名次区间Rank Spread 16 ↔ 24 确认成功Confirmed Success ▲7.66%±4.19% 赞扬/抱怨Praise vs Complaint ▲9.93%±5.32% 可控性Steerability ▲5.24%±4.67% Bash 恢复Bash Recovery ▲6.46%±1.19% 工具幻觉Tool Hallucination ▼0.99%±0.24% 会话数Sessions 6,413
› 9	小米Xiaomi Mimo V2.5 Pro · MIT	3.71% ±1.87%

综合排名Overall Rank — 名次区间Rank Spread 18 ↔ 25 确认成功Confirmed Success ▲6.18%±4.22% 赞扬/抱怨Praise vs Complaint ▲10.28%±5.58% 可控性Steerability ▲5.48%±5.30% Bash 恢复Bash Recovery ▲2.77%±2.41% 工具幻觉Tool Hallucination ▼0.63%±0.36% 会话数Sessions 6,760
› 10	SpaceXAISpaceXAI Grok 4.3 (High) · Proprietary	7.65% ±0.79%

综合排名Overall Rank — 名次区间Rank Spread 25 ↔ 29 确认成功Confirmed Success ▲8.92%±1.80% 赞扬/抱怨Praise vs Complaint ▲14.41%±2.09% 可控性Steerability ▲6.78%±1.44% Bash 恢复Bash Recovery ▲8.88%±2.03% 工具幻觉Tool Hallucination ▼0.74%±0.17% 会话数Sessions 38,049
› 11	NVIDIANvidia Nemotron 3 Ultra · OpenMDW-1.1	11.87% ±2.84%

综合排名Overall Rank — 名次区间Rank Spread 28 ↔ 32 确认成功Confirmed Success ▲11.23%±5.46% 赞扬/抱怨Praise vs Complaint ▲7.15%±9.03% 可控性Steerability ▲20.97%±5.49% Bash 恢复Bash Recovery ▲18.97%±6.56% 工具幻觉Tool Hallucination ▼1.01%±1.00% 会话数Sessions 8,814

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