GLM-4.7 Flash (Reasoning): benchmark results
Provider: Zhipu. Released 2026-01-19. Access: Open.
Unified ELO 1550 ± 1, rank #1055 of 3363 rated models, from 170 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Tau2-Bench Telecom | 98.83 | Success Rate (%) | 99.8 |
| AA TAU-2 Bench | 98.83 | Accuracy (%) | 99.5 |
| SEA-HELM (Filipino) - Summarization | 22.53 | Normalized Score | 86 |
| SEA-HELM (English) - SuperGPQA | 39.42 | Normalized Score | 73.7 |
| AA IFBench | 60.82 | Accuracy (%) | 72.2 |
| AA Omniscience - Software Engineering (SWE) - Dart | 24 | Accuracy (%) | 72.2 |
| SEA-HELM (English) - MathArena | 32.38 | Normalized Score | 71.9 |
| SEA-HELM (English) - AA-LCR | 43.26 | Normalized Score | 68.4 |
| SEA-HELM (English) - Knowledge | 61.71 | Normalized Score | 68.4 |
| AA Omniscience - Software Engineering (SWE) - Java | 18 | Accuracy (%) | 66.9 |
| SEA-HELM (Indonesian) - Global MMLU Lite | 73.97 | Normalized Score | 66.7 |
| SEA-HELM (English) - LiveCodeBench v6 | 54.31 | Normalized Score | 64.9 |
Interactive version: theaggregate.ai/model?slug=glm-4-7-flash-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-23.