Model
GPT-5.6 Luna
OpenAI · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Balrog | 45.6 | average_progress | N/A | Independent | medium | balrogai.com |
| CritPt | 20.6 | accuracy | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 91.6 | accuracy | N/A | Independent | high | epoch.ai |
| HouseholdBench | 1.0 | relmae | N/A | Independent | medium | arxiv.org |
| SEMAADB | 73.0 | f1 | N/A | Independent | medium | arxiv.org |
| SimpleQA Verified | 41.0 | accuracy | N/A | Independent | high | epoch.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ArchitectureIQ | 67.8 | accuracy | architectureiq | Independent | medium | arxiv.org |
| DeepSWE | 67.2 | pass@1 | mini-swe-agent | Independent | medium | deepswe.datacurve.ai |
| Hob-VL | 49.0 | accuracy | N/A | Independent | medium | arxiv.org |
| SciCode | 53.6 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-CC | 78.6 | resolve_rate | mini-swe-agent | Independent | medium | arxiv.org |
| TestSuiteBench | 19.3 | score | N/A | Independent | medium | arxiv.org |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| 2D-FET-Bench V2 | 62.3 | pass@1 | react-3/c0 | Independent | medium | arxiv.org |
| AgentMonBench | 0.7 | silentswap | N/A | Independent | medium | arxiv.org |
| AgentPersonaBench | 63.4 | accuracy | N/A | Independent | medium | arxiv.org |
| APTInvestBench | 43.8 | coverage_rate | N/A | Independent | medium | arxiv.org |
| BrickBench | 898.0 | elo | N/A | Independent | medium | arxiv.org |
| cua-speedrun | 81.6 | success_rate | N/A | Independent | medium | arxiv.org |
| DelegationBench | 60.0 | accuracy | N/A | Independent | medium | arxiv.org |
| HygieneRoboBench | 46.6 | resolved_rate | N/A | Independent | medium | arxiv.org |
| HyperBrowseComp | 11.6 | accuracy | N/A | Independent | medium | arxiv.org |
| IGMBench | 64.5 | success_rate | N/A | Independent | medium | arxiv.org |
| KlinikeBench | 7.5 | strict_pass@1 | N/A | Independent | medium | arxiv.org |
| OpenGameEval | 40.3 | pass_rate | N/A | Independent | medium | arxiv.org |
| ReFract | 0.6 | success_rate | function calling | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| 4MT-VLM | 45.0 | accuracy | N/A | Independent | medium | arxiv.org |
| CogEmo-Bench | 43.5 | m-f1 | N/A | Independent | medium | arxiv.org |
| EviLens | 35.5 | accuracy | N/A | Independent | medium | arxiv.org |
| GDP.pdf | 22.7 | gdp.pdf_score | N/A | Independent | medium | surgehq.ai |
| NextAct | 0.9 | h_10 | N/A | Independent | medium | arxiv.org |
| OmniVCBench | 43.7 | accuracy | N/A | Independent | medium | arxiv.org |
| PlotGround-1k | 69.7 | accuracy | N/A | Independent | medium | arxiv.org |
| SpaceCast-Bench | 46.6 | accuracy | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| APEX-Harm | 11.5 | success_rate | N/A | Independent | medium | arxiv.org |
| DistressBench | 0.7 | score | N/A | Independent | medium | arxiv.org |
| DRYLAB-Bench | 0.3 | floor-included_brs_s2 | N/A | Independent | medium | arxiv.org |
| Unified Misuse Monitoring Benchmark | 0.9 | auc | N/A | Independent | medium | arxiv.org |
Abstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…