Gemma 3 4B
Gemma / Google
33scores
31benchmarks
$0 / $0 per 1M tokenscost in/out
Metadata
Gemma Closed/API
Aliases: gemma-3-4b-it, gemma-3-4b-it:free, google-gemma-3-4b-it, google/gemma-3-4b-it, google/gemma-3-4b-it:free
| Benchmark | Category | Rank | Score | Sampled |
|---|---|---|---|---|
| Berkeley Function-Calling Leaderboard | Agentic | 101 | 19.62% | 2026-05-27 |
| LLM-WikiRace | Agentic | 27 | 2.70 | 2026-05-06 |
| Tau2-Bench Telecom | Agentic | 379 | 5% | 2026-05-11 |
| Terminal-Bench Hard | Agentic | 341 | 0.8% | 2026-05-11 |
| OpenUGI | Alignment | 1124 | 16.90 | 2026-05-06 |
| Natural2Code | Coding | 7 | 0.70 | 2026-05-06 |
| SciCode | Coding | 440 | 7.3% | 2026-05-11 |
| GSMA Open Telco Leaderboard | Domain | 75 | 39.70 | 2026-05-06 |
| Vectara HHEM Hallucination Leaderboard | Factuality | 28 | 93.60 | 2026-05-06 |
| Global-MMLU-Lite | General Knowledge | 13 | 0.55 | 2026-05-06 |
| NeedleBench | Generalization | 12 | 64.42% | 2026-05-27 |
| Artificial Analysis Intelligence Index | Intelligence | 497 | 6.3 | 2026-05-11 |
| FACTS Grounding | Intelligence | 17 | 0.30 | 2026-05-06 |
| Humanity's Last Exam | Intelligence | 276 | 5.2% | 2026-05-11 |
| MMLU-Pro | Intelligence | 315 | 41.7% | 2026-05-11 |
| La Leaderboard | Language | 10 | 25.67 | 2026-05-06 |
| Open Portuguese LLM Leaderboard | Language | 416 | 78.92 | 2026-05-06 |
| Ukrainian LLM Leaderboard | Language | 12 | 9.88 | 2026-05-06 |
| AIME 2025 | Math | 228 | 12.7% | 2026-05-11 |
| HiddenMath | Mathematics | 7 | 0.43 | 2026-05-06 |
| BRIDGE Medical Leaderboard | Medical | 125 | 38.82 | 2026-05-27 |
| BRIDGE Medical Leaderboard | Medical | 240 | 28.56 | 2026-05-27 |
| BRIDGE Medical Leaderboard | Medical | 243 | 28.19 | 2026-05-27 |
| Medmarks | Medical | 70 | 0.3214340555088692 | 2026-05-27 |
| ChartQA | Multimodal | 24 | 0.69 | 2026-05-06 |
| InfoVQA | Multimodal | 9 | 0.50 | 2026-05-06 |
| Artificial Analysis Openness Index | Openness | 52 | 50 | 2026-05-11 |
| BIG-Bench Extra Hard | Reasoning | 8 | 0.11 | 2026-05-06 |
| ECLeKTic | Reasoning | 4 | 0.05 | 2026-05-06 |
| GPQA Diamond | Reasoning | 453 | 29.1% | 2026-05-11 |
| ThaiSafetyBench | Safety | 19 | 28.11% overall ASR | 2026-05-28 |
| CritPt | Science | 194 | 0% | 2026-05-11 |
| WMT24++ | Translation | 14 | 0.47 | 2026-05-06 |
No matching rows.