OpenAI: GPT-5.4 vs Anthropic: Claude Sonnet 4.6
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| OpenAI: GPT-5.4 | Anthropic: Claude Sonnet 4.6 | |
|---|---|---|
| Vendor | openai | anthropic |
| Quality Score | 100 | 100 |
| Benchmark Score | 89.6 | 80.1 |
| Input Price | $2.50/M | $3.00/M |
| Output Price | $15.00/M | $15.00/M |
| Context Window | 1,050,000 | 1,000,000 |
| Max Output | 128,000 | 128,000 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | 84.8 | 77.9 |
| ai_index_agentic | 67.8 | 67.3 |
| ai_index_coding | 100.0 | 100.0 |
| aider_polyglot | - | 61.3 |
| eqbench | 75.6 | 68.4 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, GPT-5.4 costs $6.25 and Claude Sonnet 4.6 costs $6.75, so GPT-5.4 is slightly cheaper for the same work.
Both take in a similar amount of text: 1,050,000 and 1,000,000 tokens, about 1,575 and 1,500 pages.
On our blended benchmark score GPT-5.4 leads 89.6 to 80.1.
Across the 20 tasks where they differ, Claude Sonnet 4.6 on 20 (SQL Generation, Code Review, Code Completion and others).
GPT-5.4 was released in March 2026 and Claude Sonnet 4.6 in February 2026.
Pick GPT-5.4 if you need lower cost or benchmark performance; pick Claude Sonnet 4.6 if you need the stronger all-round task scores.
Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.
Who wins by task?
| Task | OpenAI: GPT-5.4 | Anthropic: Claude Sonnet 4.6 |
|---|---|---|
| SQL Generation | 174 | 183 |
| Code Review | 175 | 183 |
| Code Completion | 120 | 120 |
| Code Refactoring | 173 | 179 |
| Bug Fixing | 187 | 196 |
| Unit Test Generation | 158 | 166 |
| Code Documentation | 146 | 148 |
| Regex Writing | 136 | 139 |
| CI/CD Pipelines | 149 | 155 |
| Frontend Component Design | 148 | 154 |
| Data Analysis | 173 | 183 |
| CSV / Spreadsheet Cleanup | 157 | 163 |
| ETL Scripting | 161 | 167 |
| JSON Extraction | 137 | 142 |
| Bulk Data Labeling | 122 | 124 |
| OCR / Document Parsing | 148 | 152 |
| Table Extraction from PDFs | 148 | 152 |
| Long-Document Summarization | 167 | 172 |
| Short-Form Summarization | 122 | 124 |
| Blog Post Writing | 143 | 147 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →