Model
Claude Opus 4.5 (2025-11-01)
Anthropic · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Balrog | 43.5 | average_progress | N/A | Independent | medium | balrogai.com |
| CL-bench | 21.1 | overall | N/A | Independent | low | epoch.ai |
| Colombian Legal Reliability Benchmark | 86.9 | accuracy | N/A | Independent | medium | arxiv.org |
| EnigmaEval | 11.9 | accuracy | N/A | Independent | medium | labs.scale.com |
| GDPval | 45.5 | win_rate | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 80.7 | accuracy | N/A | Independent | high | epoch.ai |
| HLE | 25.2 | accuracy | N/A | Independent | low | epoch.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AptMQL-Bench | 70.3 | soft-ex | N/A | Independent | medium | arxiv.org |
| GSO | 26.5 | score_opt@1 | OpenHands | Independent | medium | gso-bench.github.io |
| SWE-bench Verified | 76.7 | resolved_rate | N/A | Independent | high | epoch.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| DeepResearch Bench | 54.8 | average_score | N/A | Independent | medium | evals.futuresearch.ai |
| OSWorld | 66.3 | score | Claude Opus 4.5 | Self-reported | medium | anthropic.com |
| Terminal-Bench | 63.1 | accuracy | Droid | Independent | medium | tbench.ai |
| Terminal-Bench | 59.1 | accuracy | Letta Code | Independent | medium | tbench.ai |
| Terminal-Bench | 58.4 | accuracy | Mux | Independent | medium | tbench.ai |
| Terminal-Bench | 57.8 | accuracy | Terminus 2 | Independent | medium | tbench.ai |
| Terminal-Bench | 54.3 | accuracy | Goose | Independent | medium | tbench.ai |
| Terminal-Bench | 52.1 | accuracy | Claude Code | Independent | medium | tbench.ai |
| Terminal-Bench | 51.9 | accuracy | OpenHands | Independent | medium | tbench.ai |
| Terminal-Bench | 51.7 | accuracy | OpenCode | Independent | medium | tbench.ai |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| GeoBench | 75.0 | acw_country_% | N/A | Independent | medium | epoch.ai |
| SQUARE-Bench | 86.7 | overall | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Cybench | 82.0 | unguided_%_solved | N/A | Self-reported | medium | assets.anthropic.com |
Abstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…