Compare / Llama 3.2 3B Instruct vs Llama 4 Scout
Llama 3.2 3B Instruct vs Llama 4 Scout
17 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Llama 3.2 3B Instruct | Llama 4 Scout |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,146 | 1,316 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,145 | 1,289 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,179 | 1,309 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,146 | 1,299 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,167 | 1,322 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,143 | 1,305 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 52.1% | 44.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 6.2% | 8.2% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 4.0% | 9.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 21.9% | 28.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 87.5% | 100.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 82.7% | 89.4% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 58.3% | 74.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 1.0% | 14.5% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 31.0% | 19.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.36 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 10.4 | 26.7 |
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.