Model
Gemma 3 27B
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| CritPt | 0.0 | accuracy | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 47.7 | accuracy | N/A | Independent | high | epoch.ai |
| LiveBench | 50.0 | global_average | N/A | Independent | medium | livebench.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Aider Polyglot | 4.9 | percent_correct | N/A | Independent | medium | aider.chat |
| SciCode | 21.2 | score | N/A | Independent | medium | artificialanalysis.ai |
| Web-Bench | 9.9 | pass@1 | N/A | Independent | medium | arxiv.org |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| GeoNatureAgent (GNA) | 15.8 | accuracy | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| GeoBench | 52.0 | acw_country_% | N/A | Independent | medium | geobench.org |
| SQUARE-Bench | 79.6 | overall | N/A | Independent | medium | arxiv.org |
| WOVEN | 62.2 | accuracy | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & AlignmentReproduced by the community
Loading community runs…