GLM-5

Z.ai llm

Z.ai's frontier flagship, roughly double the scale of GLM-4.5, built for long-horizon agentic engineering with sparse attention over a 200K-token context.

overall ⌄

51.3

54 of 125 ranked

price · $/M tokens

1 / 3.2

in / out, as of 2026-07-25

to run it yourself ⌄

446 GB

minimum · 1.8 TB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmarkpublishedmeasured
GPQA86 —
SWE-bench77.8 72.8
HLE30.5 —
BrowseComp62 —

Cheaper, and at least as good

On a 1,000-in, 1,000-out call these cost less and rank no lower — a hard comparison to argue with, and a narrow one. Value covers what it misses.

DeepSeek V4-Flash DeepSeek$0.00042overall 57.9
Claude Haiku 5.5 Anthropic$0.00060overall 65.3
Qwen3.8-Flash-Next Alibaba$0.00062overall 61.2
DeepSeek V4-Pro DeepSeek$0.00130overall 65.6
DeepSeek-V4.1-Flash DeepSeek$0.00150overall 59.2

Lineage

No recorded lineage — root or standalone model.

view in full graph →

Metadata

orgZ.ai
released2026-02-12
params744B-A40B
licensemit
hf idzai-org/GLM-5