Model evidence

GLM-5.3

Compare measured GLM-5.3 effort levels without turning one custom endpoint into a universal provider result.

Current batch

GLM-5.3 / High is both the highest-scoring measured configuration at 68/100 and the fastest in this family at 23m 35s in the selected batch.

Selected evidence
Published
Aug 26, 2026, 11:00 PM UTC
Question pack
coding-fast-v4.10
Coverage
2 measured configurations
Batch
snapshot-2026-08-26T23-00-00Z-r3

Measured configurations

ConfigurationRankScoreElapsedReference costRoute
GLM-5.3 / High#868/10023m 35s$0.42Custom endpoint
GLM-5.3 / Max#1266/10025m 6s$0.44Custom endpoint

Five-question profile

QuestionGLM-5.3 / HighGLM-5.3 / Max
Black-box Regression Audit13/2010/20
Retry Planner Counterexamples16/2014/20
CI Adversarial Audit12/2014/20
Transaction Regression Design15/2015/20
Cache Regression Test Design12/2013/20

Read High and Max as route-specific evidence

When both efforts appear in the same batch, their score and elapsed time are directly inspectable under the recorded protocol. A higher effort label still does not guarantee a higher score in every batch.

A small gap should be checked against the five-question profile and repeated comparable batches before it changes a default configuration.