Claude Opus 5
Anthropic 面向深度推理、复杂智能体编程与长周期企业任务的旗舰模型。
上下文窗口
1M tokens
claude-opus-5
最大输出
128k tokens
1M tokens
输入
$4.5$5
10% off1M tokens
输出
$22.5$25
10% off1M tokens
claude-opus-5试玩区
claude-opus-5Click or drag
暂无输出
定价
| 方案 | 上下文窗口 | 最大输出 | 输入 | 输出 | 缓存写入 | 缓存读取 |
|---|---|---|---|---|---|---|
claude-opus-5 | 1M tokens | 128k tokens | $4.5$510% off 1M tokens | $22.5$2510% off 1M tokens | $5.625$6.2510% off 1M tokens | $0.45$0.5010% off 1M tokens |
1h Cache Write · $9 · 1M tokens
缓存读取或缓存输入价格适用于复用的提示词 tokens。
README
- Claude Opus 5 面向高要求的智能体与长周期任务，包括多文件软件工程、复杂重构、代码审查、深度研究，以及需要跨越多个步骤持续推理的企业工作流。
- 该模型提供 100 万 Token 上下文窗口和 128k Token 最大输出。自适应思考默认开启，团队可通过 low 到 max 的 effort 控制，在推理深度、延迟与 Token 消耗之间取得平衡。
- 相比 Claude Opus 4.8，Claude Opus 5 在深度推理、智能体编程、长上下文指令遵循、视觉理解、文档任务和多智能体协作方面均有提升，并支持通过测试版能力在保留 Prompt 缓存的同时于对话中调整工具。
API
API
端点
POST /v1/messages认证
Authorization: Bearer YOUR_REACH_API_KEY anthropic-version: 2023-06-01 Content-Type: application/json
示例
curl -X POST "https://direct.reachapi.ai/v1/messages" \
-H "Authorization: Bearer YOUR_REACH_API_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{
"model": "YOUR_MODEL_ID",
"messages": [
{
"role": "user",
"content": "Introduce ReachAPI in one sentence."
}
],
"max_tokens": 512,
"temperature": 0.2
}'相关模型
Claude Fable 5
Anthropic's most capable widely released model for next-generation intelligence and long-running agents.
Claude Sonnet 5
Anthropic's new Sonnet model for extended-context coding, agents, and enterprise production workflows.
Claude Opus 4.8
Anthropic's newest Opus-tier model for complex reasoning, long-horizon agentic coding, and high-autonomy enterprise work.
Claude Opus 4.7
Anthropic's premium Opus model for advanced coding, vision-heavy review, and long-running enterprise agents.
Claude Opus 4.6
Anthropic's most capable model for coding, agents, and high-stakes enterprise work.
Claude Sonnet 4.6
Anthropic's high-performance daily driver for coding, agents, and professional workflows at scale.