DeepSeek V3 scores 12.4 on the Respan Index (#132), ahead of gpt-oss-20b at 11.7 (#133). gpt-oss-20b leads on 5 of the 6 benchmarks both report. gpt-oss-20b has the larger context window (131K tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 25% |
| 89.2% |
| SWE-bench Verified | 42% | 60.7% |
Higher is better. Each score links to where it was published.