GLM-5.2

Name: GLM-5.2
Author: Zhipu AI

by Zhipu AIGLM

Open WeightModel weights are publicly available — can be downloaded and self-hostedAPIAvailable through a hosted API — pay per token, no self-hosting required

Released June 20261000K context≈ 750,000 words744B/40B MoE params

A frontier-class open-weight model excelling at long-context coding tasks, competitive with the best proprietary models while being fully open under MIT license.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Coding

Best-in-class coding performance, outperforms GPT-5.5 on long-horizon benchmarks

Exceptional

GLM-5.2

by Zhipu AIGLM

Open WeightModel weights are publicly available — can be downloaded and self-hostedAPIAvailable through a hosted API — pay per token, no self-hosting required

Released June 20261000K context≈ 750,000 words744B/40B MoE params

A frontier-class open-weight model excelling at long-context coding tasks, competitive with the best proprietary models while being fully open under MIT license.

Capabilities

Capability scores are AI-generated based on model documentation, benchmarks, and technical specifications. Learn more

Coding

Best-in-class coding performance, outperforms GPT-5.5 on long-horizon benchmarks

Exceptional

Reasoning & Logic

Frontier-level reasoning, matching top proprietary models

Exceptional

Capable conversational model with deep knowledge

Strong

Broad knowledge base from large-scale pretraining

Strong

Instruction Following

Strong instruction following with 1M context window

Strong

Creative text generation across domains

Strong

Multilingual

Strong multilingual support including Chinese and English

Strong

744B total but 40B active via MoE architecture

Moderate

Standard safety alignment measures

Moderate

Use Case Fit

Fit scores are AI-generated based on model capabilities, intended use, and technical specifications. Learn more

Glossary

Frontier-ClassA model that represents the current state-of-the-art or cutting edge in AI capabilities, competing with the most advanced models available.Long-ContextThe ability of a model to process and understand very long sequences of text while maintaining coherence across distant parts of the input.MIT LicenseA permissive open-source license that allows free use, modification, and distribution of software with minimal restrictions.Open-Weight ModelA model whose trained weights are publicly released, allowing anyone to download and run it locally.