Z.ai Releases GLM-5.3-Flash with Native Multimodality and 1M Context Window

LinkZ.aiResource

Z.ai has launched GLM-5.3-Flash, available via Ollama Cloud. The model features 320B total parameters with only 18B active, delivering strong coding and agentic performance at low operational cost.

Key points

  1. Architecture features 320B total and 18B active parameters across 45 layers, cutting attention compute by 3.0× and KV cache size by 4.4× compared to GLM-5.3.
  2. Supports a 1M token context window and accepts multimodal inputs, including text, images, and video.
  3. Combines linear attention with sparse attention using IndexPool compression to minimise latency and memory overhead at long context.
  4. Hosted on Ollama Cloud under an MIT licence, priced at $0.15 per 1M input tokens, $0.03 for cached tokens, and $0.50 per 1M output tokens.
  5. Designed for agentic coding and computer-use workflows; previously tested anonymously under the codename ox-alpha.

Why it matters

The release enables software engineering teams to deploy long-context, agentic coding models at a fraction of typical frontier-model pricing without sacrificing multimodal capabilities.

Analysis

GLM-5.3-Flash targets cost-sensitive agentic and development environments. By drastically cutting active parameters and layer counts while employing a hybrid linear-sparse attention system with IndexPool compression, Z.ai maintains an efficient 1M-token context window. Benchmark figures indicate performance close to Claude Opus 4.8 on benchmarks such as DeepSWE v1.1 and the Z.ai Code Bench. The inclusion of visual coding trajectories and reinforcement learning from environment feedback specifically reinforces its utility in frontend debugging, computer use, and autonomous agent loops.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Article:docs.z.ai

    Z.AI Launches GLM-5.3-Flash Featuring Native Multimodality

    Z.AI has introduced GLM-5.3-Flash, the first native multimodal model in the GLM-5 series. Featuring 320 billion total parameters with 18 billion activated, it leverages a hybrid sparse and linear attention architecture to cut compute and memory requirements.

    AI & LLMs· Announcement

  • Link:Z.ai

    Z.ai Releases GLM-5.3-Flash as an Open-Weights Multimodal Model

    Z.ai has launched GLM-5.3-Flash under the MIT licence on Hugging Face, marking the first natively multimodal model in the GLM-5 family. Featuring 320B total parameters and 18B active parameters, it targets high efficiency across coding and agentic tasks.

    AI & LLMs· Tool

  • Link:Z.ai

    Z.ai Releases GLM-5.2 with 1M Context Window and MIT Licence

    Z.ai has released GLM-5.2 on Hugging Face under the permissive MIT licence. The flagship open-weights model is designed for long-horizon agentic and coding workflows, featuring a robust 1M-token context window. Architectural updates including IndexShare and an enhanced MTP layer significantly cut computation FLOPs and accelerate speculative decoding.

    AI & LLMs· Resource

  • Article:Z.AI

    Z.AI Quick Start: Developer Guide for GLM-5.3 and SDK Integration

    Z.AI's quick start documentation details how to connect to the platform via APIs and SDKs. In addition to showcasing the GLM-5.3 flagship model and specialized multimodal tools, it highlights full compatibility with OpenAI SDKs.

    AI & LLMs· Resource

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.