Model Comparison
OpenAI's GPT-4o beats Anthropic's Claude 3.5 Sonnet on both price and benchmarks — here's the full breakdown.
Data last updated March 4, 2026
GPT-4o is the clear winner — cheaper and higher-scoring than Claude 3.5 Sonnet. Claude 3.5 Sonnet costs $0.03 per request vs $0.0225 for GPT-4o (at 5K input / 1K output tokens). There's little reason to choose Claude 3.5 Sonnet unless you need its specific API features or ecosystem.
| Metric | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| Intelligence IndexComposite score from MMLU-Pro, GPQA, and AIME. Higher is better. | 15.9 | 17.3 |
| MMLU-ProGeneral knowledge and reasoning. Higher is better. | 0.8 | 0.8 |
| GPQAGraduate-level science questions. Higher is better. | 0.6 | 0.5 |
| AIMEMathematical problem solving. Higher is better. | 0.2 | 0.2 |
| Context windowMax tokens per request. Larger handles more text. | 200,000 | 128,000 |
List prices as published by the provider. Not adjusted for token efficiency.
| Metric | Claude 3.5 Sonnet | GPT-4o |
|---|---|---|
| Input price / 1M tokens | $3.00 | $2.50 |
| Output price / 1M tokens | $15.00 | $10.00 |
| Cache hit price / 1M tokens | $0.30 | $1.25 |
Cost per IQ point based on a typical request of 5,000 input and 1,000 output tokens.
Cheaper (list price)
GPT-4o
Higher Benchmarks
GPT-4o
Better Value ($/IQ point)
GPT-4o
Claude 3.5 Sonnet
$0.0019 / IQ point
GPT-4o
$0.0013 / IQ point
GPT-4o is 33% cheaper per request than Claude 3.5 Sonnet. GPT-4o is cheaper on both input ($2.5/M vs $3.0/M) and output ($10.0/M vs $15.0/M). The 33% price gap matters at scale but is less significant for low-volume use cases. This comparison assumes a typical request of 5,000 input and 1,000 output tokens (5:1 ratio). Actual ratios vary by workload — chat and completion tasks typically run 2:1, code review around 3:1, document analysis and summarization 10:1 to 50:1, and embedding workloads are pure input with no output tokens.
GPT-4o scores higher overall (17.3 vs 15.9). Claude 3.5 Sonnet leads on GPQA (0.6 vs 0.54) and AIME (0.16 vs 0.15), with both within 5% on MMLU-Pro. Claude 3.5 Sonnet's GPQA score of 0.6 makes it stronger for technical and scientific tasks.
Claude 3.5 Sonnet has a 56% larger context window at 200,000 tokens vs GPT-4o at 128,000 tokens. That's roughly 266 vs 170 pages of text. The extra context capacity in Claude 3.5 Sonnet matters for document analysis and long conversations.
GPT-4o offers 45% better value at $0.0013 per intelligence point compared to Claude 3.5 Sonnet at $0.0019. GPT-4o is both cheaper and higher-scoring, making it the clear value pick. You don't sacrifice quality to save money with GPT-4o.
With prompt caching, GPT-4o and Claude 3.5 Sonnet cost about the same per request. Caching saves 45% on Claude 3.5 Sonnet and 28% on GPT-4o compared to standard input prices. Claude 3.5 Sonnet benefits more from caching. If your workload has repetitive prompts, Claude 3.5 Sonnet's cache discount gives it a bigger cost advantage than list prices suggest.
Pricing verified against official vendor documentation. Updated daily. See our methodology.
Related Comparisons
Create an account, install the SDK, and see your first margin data in minutes.
See My Margin DataNo credit card required