LLM Comparison
GPT-5.5 Pro vs Claude Opus 4.6
Side-by-side specs, pricing & capabilities · Updated April 2026
Price vs Intelligence
Add to comparison
2/6 models| Organization | ||
| OpenTools Score | 97 0.9 | 63 4.2 |
| Family | GPT | Claude |
| Status | Current | Current |
| Release Date | Apr 2026 | Feb 2026 |
| Context Window | 1.1M tokens | 1.0M tokens |
| Input Price | $30.00/M tokens | $5.00/M tokens |
| Output Price | $180.00/M tokens | $25.00/M tokens |
| Pricing Notes | No cached input discount. Long context (>272K tokens): 2x input, 1.5x output. Batch API: 50% discount. Regional processing: 10% uplift. | Cache read: $0.5000/M tokens |
| Capabilities | textvisioncodetool-useextended-thinkingweb-search | textvisioncodetool-use |
| Training Cutoff | December 2025 | — |
| Max Output | 128K tokens | 128K tokens |
| API Identifier | openai/gpt-5.5-pro | anthropic/claude-opus-4.6 |
| Benchmarks | ||
| FrontierMath Tier 4 | 39.6openai | — |
| FrontierMath Tier 1-3 | 52.4openai | — |
| HLE (with tools) | 57.2openai | — |
| HLE (no tools) | 43.1openai | — |
| BrowseComp | 90.1openai | 83.7anthropic |
| GeneBench | 33.2openai | — |
| GDPval | 82.3openai | — |
| Investment Banking Modeling | 88.6openai | — |
| MMLU | — | 89.8anthropic |
| GPQA Diamond | — | 91.3anthropic |
| SWE-bench Verified | — | 80.8anthropic |
| SWE-bench Pro | — | 53.4anthropic |
| Terminal-Bench 2.0 | — | 65.4anthropic |
| MATH 500 | — | 80.7anthropic |
| LiveCodeBench | — | 55.9anthropic |
| Berkeley Function Calling | — | 91.9anthropic |
| HLE | — | 48anthropic |
| MCP-Atlas | — | 75.8anthropic |
| OSWorld-Verified | — | 72.7anthropic |
| DocVQA | — | 91.3anthropic |
| GPQA-AA Elo | — | 1619artificial-analysis |
| View GPT-5.5 Pro | View Claude Opus 4.6 | |
Cost Calculator
Enter your expected monthly token usage to compare costs.
| Model | Input | Output | Total / mo | vs Best |
|---|---|---|---|---|
| Claude Opus 4.6Cheapest | $5.00 | $12.50 | $17.50 | — |
| GPT-5.5 Pro | $30.00 | $90.00 | $120.00 | +586% |
OpenAI
GPT-5.5 Pro
GPT-5.5 Pro uses more compute to think harder and provide consistently better answers than GPT-5.5. Designed for the toughest problems, some requests may take several minutes. It excels at FrontierMath Tier 4 (39.6%) and BrowseComp (90.1%), offering the highest intelligence available from OpenAI at a premium price point.
Anthropic
Claude Opus 4.6
Claude Opus 4.6 is a multimodal LLM from Anthropic with adaptive reasoning capabilities. Achieves 91.3% on GPQA Diamond and 80.8% on SWE-bench Verified. Supports up to 1,000,000 token context window. Available from $5.00/M input tokens.
More Comparisons
Looking for more AI models?
Browse All LLMs