Anthropic: claude-opus-4.1:free
Claude Opus 4.1 is a targeted performance upgrade to Anthropic's flagship Opus 4 series. Serving as a drop-in replacement, it maintains the same API footprint and pricing while delivering substantial gains in real-world software engineering and long-horizon agentic workflows. Scoring 74.5% on SWE-bench Verified, it outperforms the original Opus 4 by providing superior multi-file refactoring and pinpoint error-tracking capabilities. It features a "hybrid reasoning" architecture that allows developers to toggle between near-instant responses and deep "Extended Thinking" modes. This makes it an industry leader for autonomous coding agents, enterprise-level knowledge assistants, and complex research-level synthesis.
價格
效能
使用量與排名
支援的參數
所有提供方:為該模型提供服務的每個上游都支援。部分提供方:取決於處理請求的上游。預設值:未設定時傳送的值。
| 參數 | 提供方 | 預設值 |
|---|---|---|
| frequency_penalty | 部分提供方 | 預設不傳送 |
| include_reasoning | 所有提供方 | - |
| max_completion_tokens | 部分提供方 | - |
| max_tokens | 所有提供方 | - |
| presence_penalty | 部分提供方 | 預設不傳送 |
| reasoning | 所有提供方 | - |
| repetition_penalty | 部分提供方 | 預設不傳送 |
| response_format | 部分提供方 | - |
| stop | 部分提供方 | - |
| structured_outputs | 部分提供方 | - |
| temperature | 所有提供方 | 預設不傳送 |
| tool_choice | 所有提供方 | - |
| tools | 所有提供方 | - |
| top_k | 部分提供方 | 預設不傳送 |
| top_p | 部分提供方 | 預設不傳送 |