Compare / Llama 3.2 1B Instruct vs Mistral Small 3.2 24B
Llama 3.2 1B Instruct vs Mistral Small 3.2 24B
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 1B Instruct | Mistral Small 3.2 24B |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,070 | 1,363 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,083 | 1,323 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,082 | 1,335 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,086 | 1,338 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,111 | 1,357 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,076 | 1,329 |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 51.6% | 87.9% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 3.2% | 18.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 0.0% | 14.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 10.8% | 37.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 43.8% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 38.4% | 89.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 11.8% | 79.0% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 0.0% | 31.0% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 22.0% | 17.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.33 | 0.35 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 11.8 | 36.3 |
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.