Claude 3.5 Sonnet
AnthropicPrevious generation Sonnet. Still excellent for coding tasks at the same price point.
Claude 3.5 Sonnet Pricing
| Token Type | Price per Million |
|---|---|
| Input tokens | $3.00 |
| Output tokens | $15.00 |
| Cache read tokens | $0.300 |
| Cache creation tokens | $3.75 |
Estimated Cost by Project Size
Realistic cost estimates for common coding scenarios. Assumes 30% cache hit rate where caching is available.
| Scenario | Token Usage | Estimated Cost |
|---|---|---|
| Small Script (1K lines) | 50K input / 30K output | $0.62 |
| Medium Feature (10K lines) | 500K input / 200K output | $4.66 |
| Large Project (50K lines) | 2,500K input / 1,000K output | $23.29 |
| Code Review (5K lines) | 250K input / 25K output | $1.20 |
Benchmark Performance — Claude 3.5 Sonnet
Third-party benchmark scores normalized to 0-100 scale. Higher is better. Aggregated scores from published third-party benchmarks. SWE-bench measures real GitHub issue resolution. LiveCodeBench measures competitive programming ability. HumanEval measures basic code generation. BigCodeBench measures practical, multi-step coding tasks. All scores normalized to 0-100 scale.
Sources: SWE-bench Verified, LiveCodeBench, HumanEval, BigCodeBench
Get Access to Claude 3.5 Sonnet
Ready to start using Claude 3.5 Sonnet? Get API access directly from Anthropic.
How Does Claude 3.5 Sonnet Compare?
| Model | Input ($/M) | Medium Feature Cost | |
|---|---|---|---|
| Claude 3.5 Sonnet | $3.00 | $4.66 | selected |
| Claude Sonnet 4 | $3.00 | $4.66 | Compare |
| Claude 3 Sonnet | $3.00 | $4.05 | Compare |
| Qwen 3.6 Plus | $3.00 | $4.66 | Compare |
| Grok 3 | $3.00 | $4.05 | Compare |
| GPT-4o | $2.50 | $3.06 | Compare |
Related Models
Claude Sonnet 4
AnthropicAnthropic's balanced model for coding and general tasks. Best price-performance ratio in the Claude family.
Claude Opus 4
AnthropicAnthropic's most powerful model. Best for complex reasoning and challenging coding tasks.
Claude 3.5 Haiku
AnthropicFast, cost-effective model for high-volume tasks. Great for code review and simple queries.
Claude 3 Opus
AnthropicFirst generation Opus. Highest reasoning capability in the Claude 3 family.