Compare / DeepSeek V4 Pro 0423 vs gpt-oss-20b
DeepSeek V4 Pro 0423 vs gpt-oss-20b
18 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | DeepSeek V4 Pro 0423 | gpt-oss-20b |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,460 | 1,323 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,446 | 1,240 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,480 | 1,314 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,452 | 1,281 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,458 | 1,318 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,451 | 1,275 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 68.8 | 20.7 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 36.0 | 10.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.8% | 68.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 41.0% | 11.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 80.3% | 34.7% |
| Artificial Analysisreported | SciCode% of problems | 51.0% | 38.9% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 78.7% | 13.9% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 14.1% | 0.0% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 39.6% | 7.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 34.0% | 98.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.31 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 68.4 | 24.6 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.