Z.ai: GLM 5.2

z-ai

flagship

Capabilities

ToolsStructured OutputExtended ThinkingLong ContextCode

Strengths

Massive 1M-token context window for long-document workflows
Open weights enable self-hosting and full data privacy control
Strong reasoning and code capabilities with extended thinking
Competitive pricing for capability level ($3/M output)

Weaknesses

Very slow inference (0.35 speed score) unsuitable for latency-critical tasks
No vision capability; text-only modality limits multimodal applications
Moderate provider transparency and sustainability disclosure
Training data and methodology less public than leading open-weight models

Pricing

Input / 1M tokens

$0.95

Output / 1M tokens

Context window

1,048.576k

Transparency

Open weights

10.0 / 10

Open training data

0.0 / 10

Open methodology

6.5 / 10

Licence openness

9.0 / 10

Provider disclosure

5.5 / 10

FMTI company score

N/A

Composite:6.5 / 10

Z.ai (Zhipu AI) releases GLM weights openly under permissive licenses. Methodology moderately disclosed; training data composition not fully public. Provider transparency is improving but less comprehensive than leading open-weight vendors.

Sustainability

Inference energy

N/A

Training footprint

N/A

Provider infrastructure

5.0 / 10

Composite:4.8 / 10

Limited public disclosure of Z.ai's green infrastructure commitments. Large model size and 1M context window suggest higher inference energy; sustainability practices not clearly documented.

Task Fitness

question answering

90%

code

90%

mathematics

50%

business communication

92%

vision

analysis

88%

extraction

89%

long-form generation

94%

research

92%

reasoning

88%

summarisation

85%

translation

80%

conversation

95%