Model
Llama-3.1-405B-Instruct
Meta · open weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| BBH | 82.9 | average | N/A | Independent | medium | arxiv.org |
| GPQA Diamond | 50.9 | accuracy | N/A | Independent | high | epoch.ai |
| HellaSwag | 89.2 | overall_accuracy | N/A | Independent | medium | arxiv.org |
| MMLU | 84.5 | exact_match | N/A | Independent | medium | crfm.stanford.edu |
| PIQA | 85.9 | score | N/A | Independent | medium | arxiv.org |
| Trivia QA | 82.7 | exact_match | N/A | Independent | medium | arxiv.org |
| WinoGrande | 89.2 | accuracy | N/A | Independent | medium | arxiv.org |
| WinoGrande | 82.2 | accuracy | N/A | Self-reported | medium | arxiv.org |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| TheAgentCompany | 7.4 | resolved_rate | N/A | Independent | medium | github.com |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Cybench | 7.5 | unguided_%_solved | N/A | Independent | medium | cybench.github.io |
Reproduced by the community
Loading community runs…