Model
GLM-5.2
Zhipu AI · open weights
Knowledge & Reasoning
Knowledge & Reasoning| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| Colombian Legal Reliability Benchmark | 85.6 | accuracy | N/A | Independent | medium | arxiv.org |
| CritPt | 20.9 | accuracy | N/A | Independent | low | epoch.ai |
| GPQA Diamond | 91.9 | accuracy | N/A | Independent | high | epoch.ai |
| MechHypoBench | 42.5 | mca_(%) | codex | Independent | medium | arxiv.org |
| SimpleQA Verified | 34.2 | accuracy | N/A | Independent | high | epoch.ai |
Coding
Coding| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| ArchitectureIQ | 47.2 | accuracy | architectureiq | Independent | medium | arxiv.org |
| DeepSWE | 43.8 | pass@1 | mini-swe-agent | Independent | medium | deepswe.datacurve.ai |
| HealBench | 32.8 | resolved_rate | mini-swe-agent | Independent | medium | arxiv.org |
| MathVis | 18.9 | pass@1 | N/A | Independent | medium | arxiv.org |
| SciCode | 50.5 | score | N/A | Independent | medium | artificialanalysis.ai |
| SWE-bench Verified | 78.7 | resolved_rate | N/A | Independent | high | epoch.ai |
| SWE-Duel | 1.2 | bradley_terry_log_strength | mini-swe-agent | Independent | medium | arxiv.org |
Agentic / Tool Use
Agentic / Tool Use| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| AutoDataBench | 59.1 | performance_of_models_trained_on_optimized_data_vs_baseline_prediction_accuracy_comparison | N/A | Independent | medium | arxiv.org |
| DyadMem | 70.6 | accuracy | N/A | Independent | medium | arxiv.org |
| KaliBench | 81.3 | total_score_(%)_avg. | N/A | Independent | medium | arxiv.org |
| Legal Research Bench | 31.2 | accuracy | N/A | Independent | medium | arxiv.org |
| Open-Endedness Bench | 0.8 | e4_no_reward_hacking | N/A | Independent | medium | arxiv.org |
Instruction Following
Instruction Following| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| UXBench Pro | 0.0 | turn_bt | N/A | Independent | medium | arxiv.org |
Safety & Alignment
Safety & Alignment| Benchmark | Score | Metric | Harness | Source | Confidence | Link |
|---|---|---|---|---|---|---|
| OverAct | 2.0 | sir | N/A | Independent | medium | arxiv.org |
| RobustReview | 0.6 | joint_stability-discrimination_profile_and_human_alignment_scores | N/A | Independent | medium | arxiv.org |
| SWE-DeceptionBench | 62.5 | active_deception_rate | N/A | Independent | medium | arxiv.org |
Abstract Reasoning
Abstract ReasoningMultilingual
MultilingualReproduced by the community
Loading community runs…