Name: 智能体模型厂商排行榜（2026-06-16）- 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	14.17% ±1.54%

综合排名Overall Rank — 名次区间Rank Spread 1 ↔ 1 确认成功Confirmed Success ▲16.48%±2.55% 赞扬/抱怨Praise vs Complaint ▲29.65%±5.72% 可控性Steerability ▲13.39%±2.94% Bash 恢复Bash Recovery ▲9.46%±1.84% 工具幻觉Tool Hallucination ▼1.86%±0.23% 会话数Sessions 16,245
› 🥈	OpenAIOpenAI GPT 5.5 (xHigh) · Proprietary	8.27% ±1.73%

综合排名Overall Rank — 名次区间Rank Spread 2 ↔ 10 确认成功Confirmed Success ▲4.42%±3.17% 赞扬/抱怨Praise vs Complaint ▲17.52%±6.44% 可控性Steerability ▲3.12%±3.44% Bash 恢复Bash Recovery ▲14.42%±1.24% 工具幻觉Tool Hallucination ▼1.86%±0.23% 会话数Sessions 10,813
› 🥉	智谱 ZAIZ.ai GLM 5.2 (Max) · MIT · SiliconFlow	4.37% ±2.48%

综合排名Overall Rank — 名次区间Rank Spread 3 ↔ 13 确认成功Confirmed Success ▲9.43%±4.52% 赞扬/抱怨Praise vs Complaint ▲14.88%±9.11% 可控性Steerability ▲6.00%±4.50% Bash 恢复Bash Recovery ▲1.69%±3.28% 工具幻觉Tool Hallucination ▼1.86%±0.23% 会话数Sessions 9,242
› 4	DeepSeekDeepSeek DeepSeek V4 Pro · MIT · SiliconFlow	0.10% ±1.41%

综合排名Overall Rank — 名次区间Rank Spread 14 ↔ 19 确认成功Confirmed Success ▲0.44%±3.07% 赞扬/抱怨Praise vs Complaint ▲0.07%±4.86% 可控性Steerability ▲2.81%±3.04% Bash 恢复Bash Recovery ▲2.93%±1.33% 工具幻觉Tool Hallucination ▼0.74%±0.32% 会话数Sessions 28,002
› 5	GoogleGoogle Gemini 3.5 Flash · Proprietary	0.04% ±1.00%

综合排名Overall Rank — 名次区间Rank Spread 14 ↔ 18 确认成功Confirmed Success ▲1.98%±2.24% 赞扬/抱怨Praise vs Complaint ▲2.29%±3.28% 可控性Steerability ▲0.23%±1.88% Bash 恢复Bash Recovery ▲2.98%±1.31% 工具幻觉Tool Hallucination ▼1.69%±0.26% 会话数Sessions 30,341
› 6	月之暗面Moonshot Kimi K2.6 · Modified MIT · Fireworks	0.50% ±1.09%

综合排名Overall Rank — 名次区间Rank Spread 14 ↔ 19 确认成功Confirmed Success ▲0.36%±2.28% 赞扬/抱怨Praise vs Complaint ▲1.70%±3.64% 可控性Steerability ▲2.82%±2.20% Bash 恢复Bash Recovery ▲0.22%±1.66% 工具幻觉Tool Hallucination ▼1.86%±0.23% 会话数Sessions 33,885
› 7	MiniMaxMiniMax Minimax M3 · Proprietary · Fireworks	2.79% ±1.70%

综合排名Overall Rank — 名次区间Rank Spread 15 ↔ 20 确认成功Confirmed Success ▲2.30%±3.91% 赞扬/抱怨Praise vs Complaint ▲9.39%±5.50% 可控性Steerability ▲7.62%±3.78% Bash 恢复Bash Recovery ▲3.52%±1.40% 工具幻觉Tool Hallucination ▼1.86%±0.23% 会话数Sessions 9,870
› 8	阿里巴巴Alibaba Qwen 3.6 Plus · Proprietary · Fireworks	4.24% ±1.20%

综合排名Overall Rank — 名次区间Rank Spread 19 ↔ 22 确认成功Confirmed Success ▲2.84%±2.44% 赞扬/抱怨Praise vs Complaint ▲5.58%±4.08% 可控性Steerability ▲8.94%±2.46% Bash 恢复Bash Recovery ▲2.42%±1.62% 工具幻觉Tool Hallucination ▼1.41%±0.56% 会话数Sessions 31,933
› 9	xAIxAI Grok Build 0.1 · Proprietary	6.20% ±1.10%

综合排名Overall Rank — 名次区间Rank Spread 20 ↔ 24 确认成功Confirmed Success ▲6.78%±2.57% 赞扬/抱怨Praise vs Complaint ▲11.49%±3.54% 可控性Steerability ▲8.64%±2.24% Bash 恢复Bash Recovery ▲1.65%±1.60% 工具幻觉Tool Hallucination ▼2.42%±0.47% 会话数Sessions 27,225
› 10	NVIDIANvidia Nemotron 3 Ultra · OpenMDW-1.1	8.65% ±4.12%

综合排名Overall Rank — 名次区间Rank Spread 20 ↔ 26 确认成功Confirmed Success ▲4.39%±7.57% 赞扬/抱怨Praise vs Complaint ▲5.54%±14.18% 可控性Steerability ▲22.66%±8.28% Bash 恢复Bash Recovery ▲12.31%±7.55% 工具幻觉Tool Hallucination ▼1.63%±0.51% 会话数Sessions 3,926

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