Model
GPT-5.6-terra
OpenAI · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Balrog | 53.2 | average_progress | N/A | Independent | medium | balrogai.com |
| CritPt | 30.0 | accuracy | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 93.3 | accuracy | N/A | Independent | high | epoch.ai |
| HouseholdBench | 0.9 | relmae | N/A | Independent | medium | arxiv.org |
| MOF-VERIFY | 62.8 | macro-f1 | N/A | Independent | medium | arxiv.org |
| SEMAADB | 77.5 | f1 | N/A | Independent | medium | arxiv.org |
| SimpleQA Verified | 43.2 | accuracy | N/A | Independent | high | epoch.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| DeepSWE | 69.6 | pass@1 | mini-swe-agent | Independent | medium | deepswe.datacurve.ai |
| HealBench | 38.1 | resolved_rate | codex | Independent | medium | arxiv.org |
| SciCode | 55.0 | score | N/A | Independent | medium | artificialanalysis.ai |
| TaoD2C-Bench | 62.5 | accuracy | N/A | Independent | medium | arxiv.org |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AgentMonBench | 0.9 | silentswap | N/A | Independent | medium | arxiv.org |
| AgentPersonaBench | 64.5 | accuracy | N/A | Independent | medium | arxiv.org |
| APTInvestBench | 43.6 | coverage_rate | N/A | Independent | medium | arxiv.org |
| BOTTLED | 0.3 | f1 | N/A | Independent | medium | arxiv.org |
| DelegationBench | 60.4 | accuracy | N/A | Independent | medium | arxiv.org |
| DUDA-Bench | 0.1 | workflow_score | react | Independent | medium | arxiv.org |
| HyperBrowseComp | 15.8 | accuracy | N/A | Independent | medium | arxiv.org |
| KlinikeBench | 8.4 | strict_pass@1 | N/A | Independent | medium | arxiv.org |
| MMPostTrainBench | 43.4 | mean_task_score_(%) | codex | Independent | medium | arxiv.org |
| OpenGameEval | 40.4 | pass_rate | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Ego-SpaCR | 38.5 | accuracy | uniform | Independent | medium | arxiv.org |
| EviLens | 43.4 | accuracy | N/A | Independent | medium | arxiv.org |
| GDP.pdf | 24.7 | gdp.pdf_score | N/A | Independent | medium | surgehq.ai |
| PlotGround-1k | 66.6 | accuracy | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AgentDoxx | 86.0 | accuracy | N/A | Independent | medium | arxiv.org |
| APEX-Harm | 35.0 | success_rate | N/A | Independent | medium | arxiv.org |
| DecepEval | 45.5 | deception_rate | N/A | Independent | medium | arxiv.org |
| DistressBench | 0.8 | score | N/A | Independent | medium | arxiv.org |
| DRYLAB-Bench | 0.2 | floor-included_brs_s2 | N/A | Independent | medium | arxiv.org |
Abstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…