Model
Claude Opus 4.7
Anthropic · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| CritPt | 12.0 | accuracy | N/A | Independent | low | epoch.ai |
| CroissantMiner | 0.7 | accuracy | single-pass | Independent | medium | arxiv.org |
| GPQA Diamond | 90.2 | accuracy | N/A | Independent | high | epoch.ai |
| HLE | 36.2 | accuracy | N/A | Independent | low | epoch.ai |
| SimpleQA Verified | 51.7 | accuracy | N/A | Independent | high | epoch.ai |
| TasteVal | 0.5 | compute_multiplier | N/A | Independent | medium | arxiv.org |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| GSO | 44.1 | score_opt@1 | OpenHands | Independent | medium | gso-bench.github.io |
| SciCode | 54.5 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-bench Verified | 83.5 | resolved_rate | N/A | Independent | high | epoch.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AgentHorizon | 76.0 | accuracy | claude code | Independent | medium | arxiv.org |
| ExploitBench | 26.5 | mean_capability | N/A | Independent | low | epoch.ai |
| MisVisAgentBench | 45.0 | pra | N/A | Independent | medium | arxiv.org |
| OpenGameEval | 45.5 | pass_rate | N/A | Independent | medium | arxiv.org |
| Synthetic Trading Benchmark | 9.0 | e0_sharpe | N/A | Independent | medium | arxiv.org |
| Terminal-Bench | 80.2 | accuracy | WOZCODE | Independent | medium | tbench.ai |
| XiangqiBench | 10.9 | pass@3 | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| FigCodeBench | 76.2 | mmos | N/A | Independent | medium | arxiv.org |
| GDP.pdf | 21.0 | gdp.pdf_score | N/A | Independent | medium | surgehq.ai |
Safety & Alignment
Safety & AlignmentAbstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…