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

BenchmarkScoreMetricPercentile
Tau2-Bench Telecom98.83Success Rate (%)99.8
AA TAU-2 Bench98.83Accuracy (%)99.5
SEA-HELM (Filipino) - Summarization22.53Normalized Score86
SEA-HELM (English) - SuperGPQA39.42Normalized Score73.7
AA IFBench60.82Accuracy (%)72.2
AA Omniscience - Software Engineering (SWE) - Dart24Accuracy (%)72.2
SEA-HELM (English) - MathArena32.38Normalized Score71.9
SEA-HELM (English) - AA-LCR43.26Normalized Score68.4
SEA-HELM (English) - Knowledge61.71Normalized Score68.4
AA Omniscience - Software Engineering (SWE) - Java18Accuracy (%)66.9
SEA-HELM (Indonesian) - Global MMLU Lite73.97Normalized Score66.7
SEA-HELM (English) - LiveCodeBench v654.31Normalized Score64.9

Interactive version: theaggregate.ai/model?slug=glm-4-7-flash-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-23.