Anthropic: Claude Sonnet 4
Analysis Summary
Claude Sonnet 4 is Anthropic's mid-2025 flagship, combining strong reasoning, coding, and agentic capability in a single model. Its coding index of 37.6 and agentic index of 47.8 are among the strongest in its intelligence tier, and its instruction-following score of 0.547 on IFBench reflects reliable compliance with structured prompts. Vision and file input are both supported alongside a 1-million-token context window.
For businesses, Claude Sonnet 4 is a strong fit for autonomous coding agents, long-document analysis, customer-facing content generation, and complex multi-step tool use. Its long-context recall score of 0.647 confirms it can maintain coherence across very long inputs, which is critical for contract review, codebase analysis, and research summarisation. The tau2 score of 0.646 indicates reliable agentic task completion.
At $3.00 input and $15.00 output per million tokens, it is priced at the premium end, rewarding high-value tasks over bulk generation. For teams already in the Anthropic ecosystem, it is a natural upgrade path that consolidates coding, agentic, and content workflows on a single, well-supported model.
Assessed June 30, 2026
Editorial notes
Claude Sonnet 4 from Anthropic delivers strong coding, agentic performance, and instruction following with a 1M token context, vision, and file input support, making it a well-rounded flagship for business workflows.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Anthropic: Claude Sonnet 4 compares
Anthropic: Claude Sonnet 4 ranks #88 of 385 AI models we track for overall intelligence, #51 of 129 for coding, #102 of 293 for agentic tasks. Its 1M-token context window is larger than 91% of the models we list. At $3.00 per million input tokens it is cheaper than 10% of comparable models.
About Anthropic: Claude Sonnet 4
Claude Sonnet 4 significantly enhances the capabilities of its predecessor, Sonnet 3.7, excelling in both coding and reasoning tasks with improved precision and controllability. Achieving state-of-the-art performance on SWE-bench (72.7%),..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Anthropic: Claude Sonnet 4 stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
anthropic/claude-sonnet-4
|
| Provider | anthropic |
| Release Date | May 22, 2025 |
| Context Length | 1,000,000 tokens |
| Max Completion | 64,000 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $3.00 | $0.003000 |
| Output | $15.00 | $0.015000 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Anthropic: Claude Sonnet 4
How much does Anthropic: Claude Sonnet 4 cost?
Anthropic: Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens.
What is the context window of Anthropic: Claude Sonnet 4?
Anthropic: Claude Sonnet 4 has a context window of 1,000,000 tokens (1M).
Is Anthropic: Claude Sonnet 4 good for coding?
On our coding benchmark index, Anthropic: Claude Sonnet 4 ranks #51 of 129 models, placing it in the broader range of the field for code generation and debugging.
What can Anthropic: Claude Sonnet 4 do?
Anthropic: Claude Sonnet 4 supports image/vision input, tool use, and function calling.
Who created Anthropic: Claude Sonnet 4?
Anthropic: Claude Sonnet 4 is developed by Anthropic and was released on May 22, 2025.
