Overview
Anthropic has released Claude 3.5 Sonnet, which achieves new state-of-the-art results on multiple benchmarks while maintaining competitive pricing.
Key Achievements
- Top Performance: Best-in-class on coding tasks
- Speed: 2x faster than Claude 3 Opus
- Cost: 60% cheaper than comparable models
- Artifacts: New feature for interactive content
Benchmark Results
| Task | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| Coding | 92.0% | 90.2% |
| Reasoning | 88.7% | 87.4% |
| Math | 78.3% | 76.6% |
Resources
Source: Anthropic