GLM-5.3

5.3 · Family: GLM

Z.ai MoE language model (753B parameters) with a 1M context window and controllable reasoning, specialized in coding and cybersecurity.

✓ Active✓ Public access⚖ Open weightsLLMReasoning modelTool-using model📁 GLM

Context window

tokens

Parameters

753B (MoE)

parameters

Max output

128,000

tokens

Release date

17 February 2026

🏢Z.aiProducer

Access:APIDownloadHostedDeployment:☁ Cloud💻 Local

Overview

GLM-5.3 is a Mixture-of-Experts language model built by Z.ai (Zhipu AI), the next iteration of the GLM family. It has 753 billion total parameters and is optimized for complex coding and long-horizon tasks.

Architecture

The model uses an MoE architecture with a DSA sparse-attention mechanism (tagged glm_moe_dsa). The context window reaches 1 million tokens and the maximum response length is 128K tokens. GLM-5.3 supports controllable reasoning via a reasoning_effort parameter with low, high and max settings.

Capabilities and results

Compared with GLM-5.2 the model delivers about a 50% improvement on Z.ai Code Bench and sets state of the art on CyberGym for vulnerability discovery, with markedly stronger exploitation capabilities. In benchmarks it scores 88.2 on Terminal-Bench 2.1, 28.3 on Terminal-Bench 3.0, 66.9 on DeepSWE, 84.5 on CyberGym and 54.4 on ExploitBench. It was released on February 17, 2026.

Classification

LLMReasoning modelTool-using model

Family: GLM

Applications

Coding AI coding assistant Cybersecurity research Autonomous agent workflows Research assistance

Access & deployment

APIDownloadHosted

CloudLocal

Weights: Open weights

Key parameters

📏 Context: 1M

🧩 Parameters: 753B (MoE)

✓ Tools · ✓ Fine-tuning

📥 Input: text, structured data, documents

Technical specification

Context window

tokens

Parameters

753B (MoE)

parameters

Max output tokens

128,000

tokens per response

License

GLM Model License (open weights) — dostęp produkcyjny przez api.z.ai / GLM Coding Plan

Hardware requirements

Open-weight 753B MoE model — self-hosting requires multi-GPU infrastructure (recommended min. 8-16x H100/H200 for 1M context handling, ~200GB+ VRAM after INT8/FP8 quantization). Z.ai serves the model commercially via api.z.ai as the main production option.

Features:✓ Tool use✓ Fine-tuning

Modalities

⬇ Input

textstructured_datadocuments

⬆ Output

textcodestructured_data

Capabilities and applications

Native model capabilities

Advanced reasoning

The ability to perform multi-step, structured reasoning: analysing problems, planning steps, and drawing conclusions from hypotheses. Reasoning-first models (e.g. GPT-5.1 Thinking) dedicate a portion of inference to chains of thought before responding.

Category: reasoning

Extended thinking mode

A reasoning-model variant with a larger inference budget: more thinking cycles, higher answer precision at the cost of response time. Choice between 'standard' and 'extended' thinking is left to the user (e.g. the selector in GPT-5.2 Pro).

Category: reasoning

Adaptive reasoning effort

The model decides how much 'thinking' to allocate to a given query: simple questions are answered quickly, complex problems receive more inference cycles. A GPT-5.1 feature (both Instant and Thinking) that shortens time on easy tasks and extends it for hard ones.

Category: reasoning

Multi-step reasoning

Carrying out multi-step chains of reasoning across long, complex tasks.

Category: reasoning

Mathematical reasoning

The model's ability to solve mathematical tasks requiring multi-step reasoning — equations, proofs, combinatorics, geometry, calculus and competition-level problems.

Category: reasoning

Coding

Generating, analysing and modifying code in many programming languages. Covers writing functions, debugging, refactoring, code review, and creating tests. Measured by benchmarks such as HumanEval and SWE-bench.

Category: coding

Agentic coding

Multi-hour, multi-step programming tasks performed autonomously by the model: cloning a repository, running tests, iterating on fixes, integrating with CLI tools. Characteristic of Codex variants (GPT-5.1-Codex-Mini, Codex-Max).

Category: coding

Agentic capability

The model's ability to autonomously plan and execute multi-step tasks by sequentially using tools, maintaining context, and adapting to intermediate results.

Category: planning

Tool use

The model's ability to call external functions, APIs and tools during a conversation: calculator, search engine, code editor, database. The model decides when and how to use a tool and interprets its result.

Category: planning

Long context

Support for large context windows — tens to hundreds of thousands (or millions) of input tokens. Enables analysis of entire codebases, long documents, and many parallel conversations without losing earlier information. GPT-5.1 supports 400,000 tokens.

Category: language

Prompt caching

Cost-performance optimisation: repeated prompt fragments (e.g. system prompt, long documentation) are cached server-side and cheaper in subsequent calls. Significantly reduces cost for applications with long contexts.

Category: other

Multilingual

Competence in many natural languages (from a few to over a hundred): understanding, generation, translation, and code-switching within a single conversation. Frontier models support a wide range of languages with comparable quality.

Category: language

MCP support

Native support for the Model Context Protocol - the model can integrate with external MCP servers, invoke their tools, and use their data sources without a dedicated wrapper.

Category: other

Application domains

Coding AI coding assistant Cybersecurity research Autonomous agent workflows Research assistance

Benchmark results

5 benchmarks

Terminal-Bench 2.0

accuracy · Terminal-Bench 2.1

88.2%