GLM-5.3-Flash

Z.ai llm

First natively multimodal GLM-5-series model, trained from a new base with hybrid sparse/linear attention; open weights under MIT.

overall ⌄

—

nothing published

price · $/M tokens

0.15 / 0.5

in / out, as of 2026-07-25

to run it yourself ⌄

192 GB

minimum · 768 GB recommended

Benchmarks

Nothing published and nothing measured. Its maker has released no figure on any benchmark this site compares, and no platform has run it.

Cheaper, and at least as good

Nothing is both cheaper and ranked as high. This model is on the value frontier — whatever you pay for it, you are paying for something.

Lineage

No recorded lineage — root or standalone model.

view in full graph →

Metadata

orgZ.ai
released2026-08-26
params320B-A18B
licensemit
hf idzai-org/GLM-5.3-Flash