Model
Gemini 3.1 Pro
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Balrog | 57.0 | average_progress | N/A | Independent | medium | balrogai.com |
| CL-bench | 20.8 | overall | N/A | Independent | low | epoch.ai |
| CL-bench Life | 16.9 | overall | N/A | Independent | low | epoch.ai |
| Colombian Legal Reliability Benchmark | 90.5 | accuracy | N/A | Independent | medium | arxiv.org |
| CritPt | 17.7 | accuracy | N/A | Independent | low | epoch.ai |
| CroissantMiner | 0.6 | accuracy | single-pass | Independent | medium | arxiv.org |
| EnigmaEval | 36.8 | accuracy | N/A | Independent | medium | labs.scale.com |
| GameCommBench | 0.5 | analytical_completeness | N/A | Independent | medium | arxiv.org |
| GeoFitBench-LLM | 77.4 | success_rate | N/A | Independent | medium | arxiv.org |
| GPQA Diamond | 94.4 | accuracy | N/A | Independent | high | epoch.ai |
| GPQA Diamond | 91.9 | accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| HLE | 46.4 | accuracy | N/A | Independent | low | epoch.ai |
| HLE | 45.8 | accuracy | N/A | Self-reported | low | lastexam.ai |
| LMArena | 1492.0 | elo | N/A | Self-reported | low | lmarena.ai |
| MMLU-Pro | 85.0 | accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| ScholarCatalyst | 0.4 | r@20 | tool-calling agent | Independent | medium | arxiv.org |
| SimpleQA Verified | 73.5 | accuracy | N/A | Independent | high | epoch.ai |
Math
Math| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AIME 2025 | 100.0 | accuracy | N/A | Self-reported | low | kaggle.com |
| EfficientOpt | 79.1 | solver_time_ratio | N/A | Independent | medium | arxiv.org |
| MATH-500 | 94.0 | accuracy | N/A | Self-reported | low | kaggle.com |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ArchitectureIQ | 66.8 | accuracy | architectureiq | Independent | medium | arxiv.org |
| DeepSWE | 11.7 | pass@1 | mini-swe-agent | Independent | medium | deepswe.datacurve.ai |
| GSO | 22.6 | score_opt@1 | OpenHands | Independent | medium | gso-bench.github.io |
| LiveCodeBench | 81.3 | pass@1 | N/A | Self-reported | low | livecodebench.github.io |
| MathVis | 22.9 | pass@1 | N/A | Independent | medium | arxiv.org |
| PLCWorld | 36.2 | success_rate | N/A | Independent | medium | arxiv.org |
| SciCode | 58.9 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-bench Verified | 78.0 | resolved_rate | N/A | Self-reported | low | swebench.com |
| TaoD2C-Bench | 57.7 | accuracy | N/A | Independent | medium | arxiv.org |
| Terminal-Bench 2.0 | 56.2 | accuracy | N/A | Self-reported | low | tbench.ai |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AgentHorizon | 77.7 | accuracy | gemini cli | Independent | medium | arxiv.org |
| AgenticBBO-Bench | 0.6 | hpo | N/A | Independent | medium | arxiv.org |
| BOTTLED | 0.4 | f1 | N/A | Independent | medium | arxiv.org |
| BrowseComp | 59.2 | accuracy | N/A | Self-reported | low | openai.com |
| DeepResearch Bench | 47.8 | average_score | N/A | Independent | medium | evals.futuresearch.ai |
| ExploitBench | 26.1 | mean_capability | N/A | Independent | low | epoch.ai |
| HyperBrowseComp | 25.5 | accuracy | N/A | Independent | medium | arxiv.org |
| Legal Research Bench | 20.8 | accuracy | N/A | Independent | medium | arxiv.org |
| LiveMACEBench | 0.9 | ras | N/A | Independent | medium | arxiv.org |
| OpenGameEval | 51.3 | pass_rate | N/A | Independent | medium | arxiv.org |
| Synthetic Trading Benchmark | 7.7 | e0_sharpe | N/A | Independent | medium | arxiv.org |
| Taste-Bench | 65.4 | accuracy | N/A | Independent | medium | arxiv.org |
| τ²-bench | 85.3 | pass^1 | N/A | Self-reported | low | github.com |
| Terminal-Bench | 80.2 | accuracy | TongAgents | Independent | medium | tbench.ai |
| Terminal-Bench | 78.4 | accuracy | Forge Code | Independent | medium | tbench.ai |
| Terminal-Bench | 78.4 | accuracy | ForgeCode | Independent | medium | tbench.ai |
| Terminal-Bench | 74.8 | accuracy | Terminus-KIRA | Independent | medium | tbench.ai |
| Terminal-Bench | 59.4 | accuracy | Gemini CLI | Independent | medium | tbench.ai |
| XiangqiBench | 38.7 | pass@3 | N/A | Independent | medium | arxiv.org |
Multimodal
Multimodal| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| DataVista | 70.0 | accuracy | N/A | Independent | medium | arxiv.org |
| DSV-Mem | 27.5 | accuracy | N/A | Independent | medium | arxiv.org |
| EgoExo-Next | 43.8 | accuracy | N/A | Independent | medium | arxiv.org |
| EgoGears | 67.7 | accuracy | N/A | Independent | medium | arxiv.org |
| EgoTools-Bench | 69.7 | egotools-bench_overall | N/A | Independent | medium | arxiv.org |
| EMBER-Bench | 26.8 | accuracy | N/A | Independent | medium | arxiv.org |
| FigCodeBench | 83.7 | mmos | N/A | Independent | medium | arxiv.org |
| GDP.pdf | 17.0 | gdp.pdf_score | N/A | Independent | medium | surgehq.ai |
| MMMU-Pro | 81.0 | accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| NextAct | 0.3 | embedding_score | N/A | Independent | medium | arxiv.org |
| OccludeBench | 78.3 | accuracy | N/A | Independent | medium | arxiv.org |
| PlotGround-1k | 49.9 | accuracy | N/A | Independent | medium | arxiv.org |
Instruction Following
Instruction Following| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| IFEval | 85.0 | prompt_strict_accuracy | N/A | Self-reported | low | artificialanalysis.ai |
| UXBench Pro | -0.1 | turn_bt | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & AlignmentAbstract Reasoning
Abstract Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Actionable Indeterminacy Benchmark | 91.5 | accuracy | N/A | Independent | medium | arxiv.org |
| ARC-AGI-2 | 77.1 | score | N/A | Independent | low | epoch.ai |
| ARC-AGI-2 | 31.1 | accuracy | N/A | Self-reported | low | arcprize.org |
Reproduced by the community
Loading community runs…