Model
GPT-5
OpenAI · closed weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Balrog | 32.8 | average_progress | N/A | Independent | medium | balrogai.com |
| CritPt | 12.6 | accuracy | N/A | Independent | low | epoch.ai |
| EnigmaEval | 10.5 | accuracy | N/A | Independent | medium | labs.scale.com |
| GDPval | 34.8 | win_rate | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 86.2 | accuracy | N/A | Independent | high | epoch.ai |
| HLE | 25.3 | accuracy | N/A | Independent | low | epoch.ai |
| SimpleQA Verified | 50.1 | accuracy | N/A | Independent | high | epoch.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Aider Polyglot | 88.0 | percent_correct | N/A | Independent | medium | aider.chat |
| GSO | 6.9 | score_opt@1 | OpenHands | Independent | medium | gso-bench.github.io |
| SciCode | 42.9 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-bench Verified | 73.6 | resolved_rate | N/A | Independent | high | epoch.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| DeepResearch Bench | 49.6 | average_score | N/A | Independent | medium | drb.futuresearch.ai |
| ST-Bench | 11.2 | composite_score | N/A | Independent | medium | arxiv.org |
| Terminal-Bench | 49.6 | accuracy | Codex CLI | Independent | medium | tbench.ai |
| Terminal-Bench | 43.8 | accuracy | OpenHands | Independent | medium | tbench.ai |
| Terminal-Bench | 35.2 | accuracy | Terminus 2 | Independent | medium | tbench.ai |
| Terminal-Bench | 33.9 | accuracy | Mini-SWE-Agent | Independent | medium | tbench.ai |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| BrainTRACE | 41.9 | accuracy | N/A | Independent | medium | arxiv.org |
| DataVista | 67.4 | accuracy | N/A | Independent | medium | arxiv.org |
| GeoBench | 81.0 | acw_country_% | N/A | Independent | medium | geobench.org |
| RateAR | 0.8 | srcc | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & AlignmentAbstract Reasoning
Abstract ReasoningMultilingual
Multilingual| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Cross-Lingual Unlearning Tensor | 94.4 | residual_access | N/A | Independent | medium | arxiv.org |
Reproduced by the community
Loading community runs…