OpenAI: GPT-5.6 Luna vs Anthropic: Claude Sonnet 5 (batch)
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| OpenAI: GPT-5.6 Luna | Anthropic: Claude Sonnet 5 (batch) | |
|---|---|---|
| Vendor | openai | anthropic |
| Quality Score | 100 | 100 |
| Benchmark Score | 89.4 | 91.4 |
| Input Price | $0.20/M | $1.00/M |
| Output Price | $1.20/M | $5.00/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 | 84.5 | 88.0 |
| ai_index_agentic | 75.2 | 77.0 |
| ai_index_coding | 100.0 | 100.0 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, GPT-5.6 Luna costs $0.50 and Claude Sonnet 5 (batch) costs $2.25, so GPT-5.6 Luna is about 4.5 times 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 Claude Sonnet 5 (batch) leads 91.4 to 89.4.
Across the 20 tasks where they differ, GPT-5.6 Luna scores higher on 7 (Code Completion, Code Documentation, Regex Writing and others), Claude Sonnet 5 (batch) on 13 (SQL Generation, Code Review, Code Refactoring and others).
GPT-5.6 Luna was released in July 2026 and Claude Sonnet 5 (batch) in June 2026.
Pick GPT-5.6 Luna if you need lower cost; pick Claude Sonnet 5 (batch) if you need benchmark performance.
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-5.6 Luna | Anthropic: Claude Sonnet 5 (batch) |
|---|---|---|
| SQL Generation | 171 | 171 |
| Code Review | 166 | 167 |
| Code Completion | 133 | 120 |
| Code Refactoring | 163 | 164 |
| Bug Fixing | 179 | 180 |
| Unit Test Generation | 154 | 154 |
| Code Documentation | 143 | 143 |
| Regex Writing | 136 | 136 |
| CI/CD Pipelines | 144 | 144 |
| Frontend Component Design | 146 | 146 |
| Data Analysis | 170 | 171 |
| CSV / Spreadsheet Cleanup | 154 | 154 |
| ETL Scripting | 154 | 154 |
| JSON Extraction | 149 | 140 |
| Bulk Data Labeling | 135 | 126 |
| OCR / Document Parsing | 146 | 146 |
| Table Extraction from PDFs | 146 | 146 |
| Long-Document Summarization | 159 | 160 |
| Short-Form Summarization | 132 | 123 |
| Blog Post Writing | 140 | 140 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →