Gemini 3.7 Flash vs GPT-5.6 Luna
| model Gemini 3.7 Flash Google | model GPT-5.6 Luna OpenAI | |
|---|---|---|
| Context | 1M | 1.1M |
| Input $/M | — | — |
| Output $/M | — | — |
| Knowledge cutoff | 2026-03 | 2026-02 |
| Available via | Gemini API (AI Studio) | OpenAI API |
| FrontierCode (score) | 43.6 | — |
| DeepSWE | 65.3% | 67.2% |
| Terminal-Bench (accuracy) | 85.8 | — |
| AutomationBench (accuracy) | 30.4 | — |
| GDP.pdf (accuracy) | 34 | — |
| GDM-MRCR (accuracy) | 97 | — |
| SWE-bench Pro | — | 62.7% |
| Terminal-Bench (success rate) | — | 84.7 |
| GPQA Diamond (accuracy) | — | 92.3 |
| FrontierMath (accuracy) | — | 78.6 |
| BrowseComp (accuracy) | — | 83.3 |