Compare / DeepSeek V3.2 vs Gemini 3.1 Flash Lite Preview
DeepSeek V3.2 vs Gemini 3.1 Flash Lite Preview
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 V3.2 | Gemini 3.1 Flash Lite Preview |
|---|---|---|---|
| Arena (formerly LMArena)reported | OverallArena rating | 1,437 | 1,433 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,423 | 1,421 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,400 | 1,413 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,453 | 1,443 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,420 | 1,407 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,425 | 1,433 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,410 | 1,420 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 44.2 | 34.7 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 21.5 | 15.6 |
| Artificial Analysisreported | GPQA Diamond% of questions | 84.0% | 82.2% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 24.6% | 17.2% |
| Artificial Analysisreported | IFBench% of instructions | 60.7% | 77.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 73.3% | 74.3% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 46.8% | 31.1% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 35.6% | 24.2% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 90.6% | 31.3% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 92.6% | 99.6% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 6.3% | 8.2% |
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.