GLM-4.7 FP8: benchmark results
FP8 precision evaluation row for Zhipu GLM-4.7. Provider: Zhipu. Released 2025-12-22. Access: Open.
Aggregate Rating 1693 ± 16, rank #209 of 1656 rated models, from 83 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Danish | 76.67 | Average Score (%) | 99.1 |
| Medmarks - MedQA | 94.79 | Score (%) | 98.9 |
| EuroEval Danish NLU | 68.61 | NLU Average Score (%) | 98.4 |
| EuroEval Danish Common Sense Reasoning | 81.97 | Common Sense Reasoning Average Score (%) | 98.3 |
| EuroEval Danish NLU - Angry Tweets | 61.64 | Matthews correlation coefficient (%) | 98.2 |
| EuroEval Danish NLU - Dansk | 70.24 | Micro F1 excluding MISC (%) | 98.1 |
| Medmarks - MedConceptsQA Easy | 99.9 | Score (%) | 96 |
| Medmarks - CareQA EN | 94.35 | Score (%) | 95.5 |
| Medmarks - Medbullets OP4 | 87.99 | Score (%) | 94.9 |
| SLR-Bench - Medium | 60 | Accuracy (%) | 94.5 |
| Medmarks - HEAD-QA v2 | 91.32 | Score (%) | 94.3 |
| Medmarks - MedConceptsQA Hard | 86.6 | Score (%) | 94.3 |
Interactive version: theaggregate.ai/model?slug=glm-4-7-fp8 · How It Works · Data refreshed daily, snapshot 2026-10-20.