← All models

GLM-4.6

Published Publisher Zhipu AIglm-4.6

Overview

GLM-4.6 is a flagship language model positioned for coding, reasoning and agentic tasks. It was released on September 30, 2025. (docs.z.ai)

Capabilities and uses

  • Accepts text input and produces text output, with a 200K-token context window and a documented maximum output of 128K tokens.
  • Supports reasoning with tool use, long-context processing, coding, search-oriented agents and writing. Documented applications include software development, office automation, translation, content creation and research workflows. (docs.z.ai)
  • The publisher makes model weights publicly available and describes local deployment using vLLM or SGLang. (z.ai)

Limitations

The documented model is text-in/text-out; do not assume it has the visual capabilities of the separate GLM-4.6V model. The context and output limits apply, and results should be reviewed before relying on generated code, research or other content. In the publisher's launch evaluation, GLM-4.6 still lagged behind Claude Sonnet 4.5 in coding. (docs.z.ai)

API pricing

Z.AI international API, USD: $0.60 per million input tokens and $2.20 per million output tokens. Cached input is $0.11 per million tokens; cached-input storage is listed as free for a limited time. These are the official text-model rates, not a Coding Plan subscription or a third-party rate. Pricing checked November 18, 2026. (docs.z.ai)

Official sources