综合性能与选型
模态: 文本 (Text)
更新频率: weekly
● 自动定时拉取
Artificial Analysis Leaderboard
针对主流商业和开源大模型提供质量、速度(Tokens/s)与价格的独立第三方评测。
💡 评测口径提示: 工程选型核心参考,包含吞吐延迟与单位成本对照。
最后核验:2026-10-26
模型最新排名快照
共收录 15 款模型。数据均保留原榜单给出的分数与分项维度。
| 名次 | 模型名称 | 统一标识 | 机构 | 核心分值 | 分项指标明细 | 备注说明 | 跨榜对比 |
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 (max with fallback) | claude-opus-5.5-max-with-fallback | Anthropic | 58 | Intelligence Index: 58 Exact Index: 57.62 Context Window: 1M Cost / Task (USD): $5.98 Median Speed: 98 tps Price ($ In/Out 1M): $4/$20 | 机构: Anthropic | 上下文: 1M | 成本: $5.98 | 速度: 98 tps | 对比 → |
| 2 | Claude Sonnet 5.5 (max with fallback) | claude-sonnet-5.5-max-with-fallback | Anthropic | 56 | Intelligence Index: 56 Exact Index: 56 Context Window: 1M Cost / Task (USD): $5.46 Median Speed: 141 tps Price ($ In/Out 1M): $2/$10 | 机构: Anthropic | 上下文: 1M | 成本: $5.46 | 速度: 141 tps | 对比 → |
| 3 | Claude Opus 5.5 (xhigh with fallback) | claude-opus-5.5-xhigh-with-fallback | Anthropic | 56 | Intelligence Index: 56 Exact Index: 55.99 Context Window: 1M Cost / Task (USD): $3.46 Median Speed: 82 tps Price ($ In/Out 1M): $4/$20 | 机构: Anthropic | 上下文: 1M | 成本: $3.46 | 速度: 82 tps | 对比 → |
| 4 | Muse Wonder Full (max) | muse-wonder-full-max | Meta | 56 | Intelligence Index: 56 Exact Index: 55.8 Context Window: 1M | 机构: Meta | 上下文: 1M | 对比 → |
| 5 | Kimi K3.1 Pro | kimi-k3.1-pro | Kimi | 55 | Intelligence Index: 55 Exact Index: 55.09 Context Window: 1M Cost / Task (USD): $2.11 Median Speed: 51 tps Price ($ In/Out 1M): $2/$10 | 机构: Kimi | 上下文: 1M | 成本: $2.11 | 速度: 51 tps | 对比 → |
| 6 | Claude Opus 5.5 (high with fallback) | claude-opus-5.5-high-with-fallback | Anthropic | 54 | Intelligence Index: 54 Exact Index: 53.58 Context Window: 1M Cost / Task (USD): $1.82 Median Speed: 80 tps Price ($ In/Out 1M): $4/$20 | 机构: Anthropic | 上下文: 1M | 成本: $1.82 | 速度: 80 tps | 对比 → |
| 7 | Claude Fable 5.1 (max with fallback) | claude-fable-5.1-max-with-fallback | Anthropic | 53 | Intelligence Index: 53 Exact Index: 53.35 Context Window: 1M Cost / Task (USD): $7.63 Median Speed: 70 tps Price ($ In/Out 1M): $10/$50 | 机构: Anthropic | 上下文: 1M | 成本: $7.63 | 速度: 70 tps | 对比 → |
| 8 | Claude Fable 5.1 (xhigh with fallback) | claude-fable-5.1-xhigh-with-fallback | Anthropic | 53 | Intelligence Index: 53 Exact Index: 53.2 Context Window: 1M Cost / Task (USD): $5.98 Median Speed: 63 tps Price ($ In/Out 1M): $10/$50 | 机构: Anthropic | 上下文: 1M | 成本: $5.98 | 速度: 63 tps | 对比 → |
| 9 | GPT-6 Astra (max) | gpt-6-astra-max | OpenAI | 53 | Intelligence Index: 53 Exact Index: 52.67 Context Window: 1M Cost / Task (USD): $3.26 Median Speed: 67 tps Price ($ In/Out 1M): $10/$50 | 机构: OpenAI | 上下文: 1M | 成本: $3.26 | 速度: 67 tps | 对比 → |
| 10 | Gemini 4 Argon (high) | gemini-4-argon-high | 53 | Intelligence Index: 53 Exact Index: 52.56 Context Window: 1M Cost / Task (USD): $1.99 Median Speed: 81 tps Price ($ In/Out 1M): $2/$10 | 机构: Google | 上下文: 1M | 成本: $1.99 | 速度: 81 tps | 对比 → | |
| 11 | Muse Wonder (max) | muse-wonder-max | Meta | 52 | Intelligence Index: 52 Exact Index: 52.39 Context Window: 1M Cost / Task (USD): $4.54 Median Speed: 190 tps Price ($ In/Out 1M): $4/$20 | 机构: Meta | 上下文: 1M | 成本: $4.54 | 速度: 190 tps | 对比 → |
| 12 | GPT-6 Astra (xhigh) | gpt-6-astra-xhigh | OpenAI | 52 | Intelligence Index: 52 Exact Index: 52.39 Context Window: 1M Cost / Task (USD): $2.31 Median Speed: 68 tps Price ($ In/Out 1M): $10/$50 | 机构: OpenAI | 上下文: 1M | 成本: $2.31 | 速度: 68 tps | 对比 → |
| 13 | Claude Sonnet 5.5 (xhigh with fallback) | claude-sonnet-5.5-xhigh-with-fallback | Anthropic | 52 | Intelligence Index: 52 Exact Index: 51.9 Context Window: 1M Cost / Task (USD): $2.01 Median Speed: 110 tps Price ($ In/Out 1M): $2/$10 | 机构: Anthropic | 上下文: 1M | 成本: $2.01 | 速度: 110 tps | 对比 → |
| 14 | GPT-6.1 Sol (max) | gpt-6.1-sol-max | OpenAI | 52 | Intelligence Index: 52 Exact Index: 51.83 Context Window: 1M Cost / Task (USD): $0.72 Median Speed: 66 tps Price ($ In/Out 1M): $2/$10 | 机构: OpenAI | 上下文: 1M | 成本: $0.72 | 速度: 66 tps | 对比 → |
| 15 | Claude Opus 5.5 (medium with fallback) | claude-opus-5.5-medium-with-fallback | Anthropic | 51 | Intelligence Index: 51 Exact Index: 51.24 Context Window: 1M Cost / Task (USD): $1.34 Median Speed: 82 tps Price ($ In/Out 1M): $4/$20 | 机构: Anthropic | 上下文: 1M | 成本: $1.34 | 速度: 82 tps | 对比 → |
快照采集说明: 直接实时抓取自 Artificial Analysis 官方榜单页面 (https://artificialanalysis.ai/leaderboards/models) 状态流,收录 266 个模型与测试版本。评分以官方 Artificial Analysis Intelligence Index 为基准。
📊 本榜单指标与评价方法说明
选型决策建议:结合具体应用场景(如代码辅助、企业知识库检索、对话推理)参考专有指标,切勿以单一数值判定模型整体优劣。
Intelligence Index
对应 Artificial Analysis Leaderboard 评测基准中的测量项目。具体测试集构建、评测提示词与评分规则详见 官方文档 。
Context Window
对应 Artificial Analysis Leaderboard 评测基准中的测量项目。具体测试集构建、评测提示词与评分规则详见 官方文档 。
Cost / Task (USD)
对应 Artificial Analysis Leaderboard 评测基准中的测量项目。具体测试集构建、评测提示词与评分规则详见 官方文档 。
Median Speed
对应 Artificial Analysis Leaderboard 评测基准中的测量项目。具体测试集构建、评测提示词与评分规则详见 官方文档 。