GPT-5.6 Luna vs Qwen3.8 27B
| model GPT-5.6 Luna OpenAI | model Qwen3.8 27B Alibaba Qwen | |
|---|---|---|
| Context | 1.1M | 262K |
| Input $/M | — | — |
| Output $/M | — | — |
| Knowledge cutoff | 2026-02 | — |
| Available via | OpenAI API | weights only (no API yet) |
| SWE-bench Pro | 62.7% | — |
| Terminal-Bench (success rate) | 84.7 | — |
| DeepSWE | 67.2% | — |
| GPQA Diamond (accuracy) | 92.3 | — |
| FrontierMath (accuracy) | 78.6 | — |
| BrowseComp (accuracy) | 83.3 | — |