Compare / Claude Sonnet 4.5 vs Gemma 3 12B
Claude Sonnet 4.5 vs Gemma 3 12B
19 results from 4 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Sonnet 4.5 | Gemma 3 12B |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,461 | 1,386 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,454 | 1,332 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,498 | 1,274 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,464 | 1,321 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,457 | 1,342 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,454 | 1,313 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 95.0% | 70.3% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 65.0% | 27.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 61.4% | 5.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 73.2% | 30.4% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 68.8% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 88.7% | 79.4% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 81.1% | 74.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 81.0% | 4.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 5.0% | 9.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.41 | 0.29 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 66.9 | 29.9 |
| Vectara Hallucination Leaderboardreported | Answer rate% of documents | 95.6% | 97.4% |
| Vectara Hallucination Leaderboardreported | Hallucination rate% of summaries | 12.0% | 4.4% |
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.