GPT-5 mini and Qwen3.5 27B tie on the Respan Index at 18.3. Qwen3.5 27B leads on 5 of the 6 benchmarks both report. GPT-5 mini is 1.2x cheaper per token at list price. GPT-5 mini has the larger context window (400K tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 1183 |
| 1216 |
| SWE-bench Verified | 71% | 75% |
Higher is better. Each score links to where it was published.