Back to Models
Claude 3.5 Sonnet
Anthropictext9.6 / 10 Overall Rating
Anthropic's most capable model — excels at analysis, coding, and nuanced writing.
Context Window
200K
Knowledge Cutoff
2025
Max Output
16K
License
Proprietary
Where it Excels (Pros)
- 200K context window
- Best-in-class analysis
- Nuanced writing style
- Strong safety alignment
Limitations (Cons)
- No native image generation
- Limited tool use compared to GPT-4o
- Text-only input
Benchmark Breakdown
Reasoning (MMLU)94%
Coding (HumanEval)92%
Mathematics (MATH)90%
Long-Context Retrieval95%
Pricing Matrix
Input cost / 1M tokens$3.00
Output cost / 1M tokens$15.00
Compare Claude 3.5 Sonnet with other frontier models: