GLM-5.2

GLM-5.2

**GLM-5.2** is Z-AI's latest flagship model, purpose-built for long-horizon tasks that require sustained reasoning, planning, and execution across extended contexts. It represents a substantial leap in long-horizon task capability over its predecessor **GLM-5.1**, and — for the first time in the GLM series — delivers that capability on a **solid 1 million token context window**. ### Key Capabilities - **1M token context**: Reliably process and reason across massive documents, entire codebases, or long multi-turn conversations without losing coherence. - **Long-horizon task execution**: Optimized for complex, multi-step workflows where the model must maintain state, track dependencies, and follow through on extended plans. - **Structured outputs**: Native support for JSON mode, enabling reliable structured data generation for downstream applications. - **Function calling**: Built-in tool-use support for agentic workflows, API integrations, and multi-step automations. - **FP4 precision**: Deployed with efficient FP4 quantization for high throughput while preserving output quality. ### What It's Best For GLM-5.2 is ideal for teams building agentic systems, long-document analysis pipelines, code assistants operating over large repositories, research and legal review workflows, and any application that demands both a very large context window and dependable execution over long task horizons.

Explore similar models

Z.ai

GLM 5.1

GLM-5.1 is Z-AI's next-generation flagship model for agentic engineering.

$1.05/MTok

$3.50/MTok

200K

Z.ai

GLM 5

GLM-5 is a 744B-parameter open-weight frontier model from Z.ai, built for complex reasoning, coding, and long-horizon agentic tasks — and trained entirely on domestic Chinese hardware.

$0.80/MTok

$2.56/MTok

200K

Z.ai

GLM 4.7

A powerful 358B-parameter open-source model built for agentic coding, complex reasoning, and reliable tool use across multiple languages.

$0.40/MTok

$1.75/MTok

131.1K

Google

Gemini 3.1 Pro

Google's frontier reasoning model delivering enhanced software engineering, agentic reliability, and multimodal intelligence across a 1M-token context window.

$2.00/MTok

$12.00/MTok

1.0M

Anthropic

Claude 5 Sonnet

Anthropic's next-generation Sonnet model — the best combination of speed and intelligence, featuring a 1M token context window, adaptive thinking by default, and major gains in coding and agentic tasks.

$3.00/MTok

$15.00/MTok

1M

OpenAI

GPT‑5.2 Pro

The most capable model series yet for professional knowledge work.

$21.00/MTok

$168.00/MTok

400K