Compare / Qwen3 32B vs Gemini 2.5 Flash
Qwen3 32B vs Gemini 2.5 Flash
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 | Qwen3 32B | Gemini 2.5 Flash |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,340 | 1,399 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,304 | 1,394 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,398 | 1,424 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,331 | 1,400 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,347 | 1,409 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,308 | 1,402 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 82.4% | 93.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 26.7% | 41.3% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 47.9% | 36.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 48.7% | 56.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 93.8% | 75.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 90.3% | 88.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 82.0% | 78.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 26.0% | 62.0% |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 99.9% | 99.0% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 5.9% | 7.8% |
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.