Claude Sonnet 5
ClaudeFlagship
ThinkingTool UseVisionStructured Output
About this model
Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...
Performance Tier
Flagship
Claude Sonnet 5 is a flagship model from Claude : the most capable in their lineup.
Best-in-class model from this provider. Highest performance across benchmarks, ideal for demanding tasks.
Pricing
This model is included in Elosia plansHigh
High pricing. Reserve this level for tasks that demand maximum quality.
| Type | per 1M tokens |
|---|---|
| Input (prompt) | $2.00 |
| Output (completion) | $10.00 |
| Cache read | $0.200 |
| Cache write | $2.50 |
Capabilities
Context Length1.0M
Max Output Tokens128K
TokenizerClaude
Inputtext, image, file
Outputtext
Release DateJune 30, 2026
Benchmarks
General Intelligence
MMLU
Not reported
GPQA Diamond
Not reported
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
SWE-bench Verified
85.2%
SWE-bench Multilingual
78.3%
Reasoning
IFEval
Not reported
Humanity's Last Exam
57.4%
Agentic
SWE-bench Pro
63.2%
Terminal-Bench 2.1
80.4%
Where does Claude Sonnet 5 stand?
Compare its performance index against every other model.
Recommended Use Cases
CodingAnalysisResearchGeneral Chat
Strengths
- State-of-the-art agentic coding (SWE-bench Verified 85.2%, Terminal-Bench 2.1 80.4%)
- Excels at computer use and agentic search (OSWorld-Verified 81.2%, BrowseComp 84.7%)
- Frontier tool-use reasoning (Humanity's Last Exam 57.4% with tools)
- 1M token context window for long-horizon agentic tasks
Limitations
- Anthropic's published suite is agentic-focused — classic academic benchmarks (GPQA, MMLU, AIME) are not reported
- Premium cost/latency vs Haiku for simple, non-agentic queries
Frequently asked questions
Resources
This model may use your data for training