anthropic 的网站图标

Anthropic: Claude Sonnet 4.5 (batch)

anthropic/claude-sonnet-4.5:batch

Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.

Sonnet 4.5 also introduces stronger agentic capabilities, including improved tool orchestration, speculative parallel execution, and more efficient context and memory management. With enhanced context tracking and awareness of token usage across tool calls, it is particularly well-suited for multi-context and long-running workflows. Use cases span software engineering, cybersecurity, financial analysis, research agents, and other domains requiring sustained reasoning and tool use.

模态
输入/输出价格
US$1.50 / US$7.50每 1M
上下文
1.0M
发布时间
2025年9月29日
知识截止日期
2025年1月

提供商

不同公司托管同一模型。OpenRouter 根据您选择的路由模式将请求路由到其中一家公司:Balanced(价格 + 速度)、Nitro(速度最快)或 Floor(价格最低)。

定价

客户为此模型实际支付的平均价格,与提供商公布的价格并列展示。由于缓存和折扣,实际支付的价格往往远低于标价。

加权平均值

输入
US$0/百万 token
输出
US$0/百万 token
价格历史
当前选择无实际价格数据

性能

吞吐量是模型生成内容的速度(每秒 token 数,越高越好)。延迟是完整的往返时间(越低越好)。TTFT 是首个 token 的响应时间,即您看到内容出现前需要等待的时间(越低越好)。

AutoExacto 基准测试
GPQA DiamondTAU-BenchAnthropic81.7%76.2%auto-routing82.0%73.9%Amazon Bedrock81.5%73.3%Amazon Bedrock81.1%73.4%Claude Platform on AWS80.2%73.9%Azure81.5%72.5%
+1 家其他提供商

正常运行时间

正常运行时间是指过去 3 天中至少有一家提供商响应请求的时间百分比。可用性是指成功提供推理服务的时间百分比。OpenRouter 会持续监控,并在某家提供商返回错误时使用次优提供商。

正常运行时间数据尚不足,暂时无法显示。

当上游提供商发生错误时,如果您的请求筛选条件允许,我们可以通过路由到另一家健康的提供商来恢复服务。您可以通过以下接口以编程方式访问各提供商的正常运行时间数据: 端点 API。了解更多有关我们的 负载均衡和提供商路由选项。

基准测试

标准化评测得分,以及此模型针对共用基准测试提示词的实际输出。百分比越高越好,每项排名显示此模型在该来源所评测的模型中的位置。

基准测试得分摘要: Anthropic: Claude Sonnet 4.5 (batch) (Artificial Analysis、Design Arena 和 Vals AI)
来源基准测试得分
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) Coding Index52.1
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) Agentic Index15.8
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) GPQA Diamond83.4%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) HLE17.8%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) IFBench57.3%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) τ²-Bench Telecom78.1%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) AA-LCR72.3%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) τ³-Banking24.5%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) GDPval-AA20.7%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) CritPt1.1%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) LiveCodeBench71.4%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) SciCode45.7%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) Terminal-Bench Hard35.6%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) Terminal-Bench 4.00.0%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) Terminal-Bench 2.155.8%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) MMLU-Pro87.5%
Artificial AnalysisClaude 4.5 Sonnet (Reasoning) AIME 202588.0%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) GPQA Diamond72.7%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) HLE7.2%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) IFBench42.7%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) τ²-Bench Telecom70.5%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) AA-LCR54.0%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) CritPt0.0%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) LiveCodeBench59.0%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) Terminal-Bench Hard28.8%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) MMLU-Pro86.0%
Artificial AnalysisClaude 4.5 Sonnet (Non-reasoning) AIME 202537.0%
Design ArenaClaude Sonnet 4.5 Models Arena 3D Elo1137
Design ArenaClaude Sonnet 4.5 Models Arena Asciiart Elo1158
Design ArenaClaude Sonnet 4.5 Models Arena Code Categories Elo1171
Design ArenaClaude Sonnet 4.5 Models Arena Data Visualization Elo1163
Design ArenaClaude Sonnet 4.5 Models Arena Game Development Elo1155
Design ArenaClaude Sonnet 4.5 Models Arena SVG Elo1097
Design ArenaClaude Sonnet 4.5 Models Arena UI Component Elo1150
Design ArenaClaude Sonnet 4.5 Models Arena Website Elo1180
Design ArenaClaude Sonnet 4.5 Agents Arena Full Stack Elo1024
Design ArenaClaude Sonnet 4.5 Agents Arena Mobile Apps Elo1129
Design ArenaClaude Sonnet 4.5 Agents Arena Webapps Elo1040
Vals AIClaude Sonnet 4.5 (Nonthinking) CorpFin v260.8%
Vals AIClaude Sonnet 4.5 (Nonthinking) Finance Agent v149.2%
Vals AIClaude Sonnet 4.5 (Nonthinking) MedCode40.6%
Vals AIClaude Sonnet 4.5 (Nonthinking) MedScribe84.5%
Vals AIClaude Sonnet 4.5 (Nonthinking) SAGE v132.9%

应用

向此模型发送最多流量的公开应用。这能很好地反映实际生产工作负载的情况,也能提示此模型最适合哪些用例。

使用此模型的应用尚无足够数据。

活动

此模型的 token 用量和请求流量随时间的变化。

API

可直接用于调用此模型的代码。OpenRouter 的 API 与 OpenAI 兼容——大多数 SDK 只需替换基础 URL 即可使用。切换模型时,唯一需要更改的是下方的模型标识符。

目前无法通过 EU 区域内路由使用。 区域内路由文档

常见问题