Using the API
Use Chat completion API to call Claude models. See the pricing page for pricing details.Claude models comparison from Anthropic
| Feature | Claude 4 Sonnet | Claude 4 Opus | Claude 3.7 Sonnet | Claude 3.5 Sonnet | Claude 3.5 Haiku |
|---|---|---|---|---|---|
| Description | Most advanced Claude model | Highly capable for complex tasks | Enhanced Claude 3.5 Sonnet | Versatile and intelligent | Fast and efficient |
| Strengths | Top-level intelligence and reasoning | Exceptional fluency and understanding | Improved performance and efficiency | Balanced performance | Quick and accurate |
| Multilingual | Yes | Yes | Yes | Yes | Yes |
| Vision | Yes | Yes | No | No | No |
| Tool calls | Yes | Yes | Yes | Yes | Yes |
| API model name | claude-sonnet-4-20250514claude-sonnet-4-20250514-thinking | claude-opus-4-20250514claude-opus-4-20250514-thinking | claude-3-7-sonnet-latestclaude-3-7-sonnet-20250219 | claude-3-5-sonnet-20241022claude-3-5-sonnet-20240620claude-3-5-sonnet-latest | claude-3-5-haiku-20241022claude-3-5-haiku-latest |
| Comparative latency | Fastest | Fast | Fast | Moderately fast | Fastest |
| Context window | 200K | 200K | 200K | 200K | 200K |
| Max output | 8192 tokens | 4096 tokens | 8192 tokens | 8192 tokens | 8192 tokens |
| Training data cut-off | May 2025 | May 2025 | June 2024 | June 2024 | June 2024 |