Compare / DeepSeek V3.1 vs GPT-5.6 Terra
DeepSeek V3.1 vs GPT-5.6 Terra
16 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 V3.1 | GPT-5.6 Terra |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,417 | 1,466 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,402 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,432 | 1,504 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,416 | 1,462 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,417 | 1,465 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,407 | 1,445 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 13.7 | 42.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 77.9% | 92.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 14.3% | 42.9% |
| Artificial Analysisreported | IFBench% of instructions | 41.5% | 71.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 56.7% | 83.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 25.0% | 62.9% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 37.4% | 86.3% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 30.0% | 22.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.39 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 51.9 | 60.4 |
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.