Z AI
GLM-5.3 (low)
GLM 5.3 delivers comprehensive advancements in complex software engineering and agent capabilities. It uses the same base model as GLM-5.2, with all improvements driven by post-training.
Z AI releasesModel Snapshot
TypeLanguage
ReleaseAug 18, 2026
Context1,048,575
Tagsimplicit-caching, reasoning, tool-use, structured-output
Overall
Rank #93 across 701
34.3
Coding
Rank #36 across 344
74.8
Blended Price
Rank #285 across 426
$2.15/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed89.3 tok/s
First Token2.41s
Max Output943,717
Pricing
Blended$2.15/M
Input$1.40/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 34.3 | #93 |
| Artificial Analysis Coding Index | coding | 74.8 | #36 |
| Humanity's Last Exam | reasoning | 36.6% | #93 |
| SciCode | coding, reasoning | 42.0% | #138 |
| Output Speed |