Gemma 3 12B
Gemma / Google
37scores
35benchmarks
$0 / $0 per 1M tokenscost in/out
Metadata
Gemma Closed/API
Aliases: gemma-3-12b-it, gemma-3-12b-it:free, google-gemma-3-12b-it, google/gemma-3-12b-it, google/gemma-3-12b-it:free
| Benchmark | Category | Rank | Score | Sampled |
|---|---|---|---|---|
| Berkeley Function-Calling Leaderboard | Agentic | 66 | 30.43% | 2026-05-27 |
| LLM-WikiRace | Agentic | 18 | 22.70 | 2026-05-06 |
| Tau2-Bench Telecom | Agentic | 369 | 10.8% | 2026-05-11 |
| Terminal-Bench Hard | Agentic | 340 | 0.8% | 2026-05-11 |
| OpenUGI | Alignment | 1057 | 20.89 | 2026-05-06 |
| Natural2Code | Coding | 4 | 0.81 | 2026-05-06 |
| SciCode | Coding | 387 | 17.4% | 2026-05-11 |
| GSMA Open Telco Leaderboard | Domain | 53 | 46.38 | 2026-05-06 |
| Vectara HHEM Hallucination Leaderboard | Factuality | 7 | 95.60 | 2026-05-06 |
| Global-MMLU-Lite | General Knowledge | 7 | 0.69 | 2026-05-06 |
| NeedleBench | Generalization | 5 | 75.31% | 2026-05-27 |
| GeoRC | Geospatial | 11 | 31.21 | 2026-05-27 |
| Artificial Analysis Intelligence Index | Intelligence | 452 | 8.79 | 2026-05-11 |
| FACTS Grounding | Intelligence | 16 | 0.31 | 2026-05-06 |
| Humanity's Last Exam | Intelligence | 324 | 4.8% | 2026-05-11 |
| MMLU-Pro | Intelligence | 266 | 59.5% | 2026-05-11 |
| Open Portuguese LLM Leaderboard | Language | 63 | 84.97 | 2026-05-06 |
| Ukrainian LLM Leaderboard | Language | 8 | 7.82 | 2026-05-06 |
| J1-ENVS | Legal | 10 | 50.42 | 2026-05-26 |
| LEXam | Legal | 24 | 41.29% open / 29.94% MCQ | 2026-05-28 |
| AIME 2025 | Math | 213 | 18.3% | 2026-05-11 |
| HiddenMath | Mathematics | 4 | 0.55 | 2026-05-06 |
| BRIDGE Medical Leaderboard | Medical | 32 | 47.73 | 2026-05-27 |
| BRIDGE Medical Leaderboard | Medical | 145 | 37.32 | 2026-05-27 |
| BRIDGE Medical Leaderboard | Medical | 168 | 35.37 | 2026-05-27 |
| Medmarks | Medical | 57 | 0.4378666803080891 | 2026-05-27 |
| FLORES European Languages Leaderboard | Multilingual | 2 | 50.03 | 2026-05-06 |
| ChartQA | Multimodal | 23 | 0.76 | 2026-05-06 |
| IDP Leaderboard | Multimodal | 27 | 0 | 2026-05-06 |
| InfoVQA | Multimodal | 8 | 0.65 | 2026-05-06 |
| Artificial Analysis Openness Index | Openness | 49 | 50 | 2026-05-11 |
| BIG-Bench Extra Hard | Reasoning | 6 | 0.16 | 2026-05-06 |
| ECLeKTic | Reasoning | 3 | 0.10 | 2026-05-06 |
| GPQA Diamond | Reasoning | 416 | 34.9% | 2026-05-11 |
| ThaiSafetyBench | Safety | 15 | 20.40% overall ASR | 2026-05-28 |
| CritPt | Science | 190 | 0% | 2026-05-11 |
| WMT24++ | Translation | 11 | 0.52 | 2026-05-06 |
No matching rows.