GLM-4.7 FP8 — benchmark results
FP8 precision evaluation row for Zhipu GLM-4.7. Provider: Zhipu. Released 2025-12-22. Access: Open.
Unified ELO 1693 ± 19, rank #284 of 1776 rated models, from 71 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Danish | 76.67 | Average Score (%) | 99.1 |
| Medmarks - LongHealth Task 1 | 91.17 | Score (%) | 98.6 |
| Medmarks - MedQA | 94.79 | Score (%) | 98.6 |
| 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 | Sentiment classification Score (%) | 98.2 |
| EuroEval Danish NLU - Dansk | 70.24 | Named entity recognition Score (%) | 98.1 |
| Medmarks - MedXpertQA Reasoning | 43.69 | Score (%) | 95.7 |
| Medmarks - MedXpertQA Understanding | 42.11 | Score (%) | 95.7 |
| Medmarks - Medbullets OP4 | 87.99 | Score (%) | 95.7 |
| Medmarks - Medbullets OP5 | 84.52 | Score (%) | 95.7 |
| Medmarks - SuperGPQA Medicine Hard | 48.39 | Score (%) | 95.7 |
Interactive version: theaggregate.ai/model?slug=glm-4-7-fp8 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.