Model
GPT-5.4-mini
OpenAI · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| CritPt | 10.0 | accuracy | N/A | Independent | low | epoch.ai |
| CroissantMiner | 0.6 | accuracy | single-pass | Independent | medium | arxiv.org |
| GPQA Diamond | 86.9 | accuracy | N/A | Independent | high | epoch.ai |
| SimpleQA Verified | 29.4 | accuracy | N/A | Independent | high | epoch.ai |
| TPBench | 0.9 | accuracy | N/A | Independent | medium | arxiv.org |
Math
MathCoding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Closed-loop evaluation benchmark for embedded coding agents | 53.3 | pass_rate | N/A | Independent | medium | arxiv.org |
| SciCode | 49.9 | score | N/A | Independent | medium | artificialanalysis.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AgentHorizon | 69.2 | accuracy | codex | Independent | medium | arxiv.org |
| DeepResearch Bench | 36.3 | average_score | N/A | Independent | medium | evals.futuresearch.ai |
| KlinikeBench | 15.9 | strict_pass@1 | N/A | Independent | medium | arxiv.org |
| Legal Research Bench | 12.6 | accuracy | N/A | Independent | medium | arxiv.org |
| UserProxyBench | 0.7 | ufs | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| FigCodeBench | 75.2 | mmos | N/A | Independent | medium | arxiv.org |
| Front2Back-ReID | 58.8 | accuracy | N/A | Independent | medium | arxiv.org |
Long Context
Long ContextSafety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| BazaarBench | 2.0 | failure_rate | N/A | Independent | medium | arxiv.org |
Abstract Reasoning
Abstract ReasoningReproduced by the community
Loading community runs…