Anthropic Releases Claude Opus 4.5
Anthropic launched Claude Opus 4.5 on November 24, 2025. This release features a 200k context window and improvements in vision, coding, and computer use.
Release Details and Specifications
Anthropic released Claude Opus 4.5 on November 24, 2025. This date marks the official announcement of the model within the Claude family. The model identifier observed in the API is claude-opus-4-5-20251101. The release is described as combining maximum capability with practical performance, targeting complex specialized tasks, professional software engineering, and advanced agents.
- Input modalities include text and image.
- Output modality is text.
- The context window is 200,000 tokens.
- The maximum output length is 64,000 tokens.
- Open weights are not confirmed for this model.
Documented Capabilities
The source material states that Claude Opus 4.5 features step-change improvements in vision, coding, and computer use. These areas are highlighted as key differentiators for the new release. The documentation describes the model as ideal for professional software engineering workflows. No specific benchmark scores or quantitative performance metrics are provided in the available facts to quantify these improvements relative to previous iterations.
Evaluation Procedure for Engineers
To evaluate the model's suitability for specific workflows, engineers should design targeted tests based on their application requirements. Since the source confirms improvements in vision, coding, and computer use, test cases should focus on these domains. For example, provide image inputs containing code snippets and measure the accuracy of the generated explanations. Alternatively, present complex coding tasks and compare the output against known correct solutions. Validate the model's ability to handle long-context scenarios by utilizing the full 200,000-token window. Measure latency and token usage during these tests to assess practical performance. Compare results against baseline models if available in your environment to determine if the claimed improvements align with your specific use case.
Artificial Analysis benchmark review
Benchmark review updated on . These scores apply to the linked artifact and recorded methodology; they do not establish the performance of a different runtime or quantized model.
| Measurement | Recorded result | Measured |
|---|---|---|
| Artificial Analysis Intelligence Index (v4.1) ↗ | 35.6 index_points | 2026-08-09 |
| Artificial Analysis Intelligence Index (v4.2) ↗ | 27.8 index_points | 2026-09-05 |
| Artificial Analysis Intelligence Index (v4.3) ↗ | 23.7 index_points | 2026-09-09 |
Review the full benchmark provenance and source methodology →
Primary sources
Frequently asked questions
What is the context window size for Claude Opus 4.5?
Claude Opus 4.5 has a context window of 200,000 tokens. The maximum output length is set at 64,000 tokens.
Are open weights available for this model?
The source indicates that open weights are not confirmed for Claude Opus 4.5. This does not establish whether local installation is impossible or mandatory, only that open weights have not been verified as available.