Roar AI
DiscordGet started
All models
Zhipu AI · Open weights

GLM-5.3

GLM-5.3 costs $1.274 per million input tokens and $4.004 per million output tokens on Roar AI. Z.ai’s open flagship for coding agents and security work.

Input$1.274per 1M tokens
Output$4.004per 1M tokens
Context window1Mtokens

GLM-5.3 API pricing

Input$1.274 per 1M tokens
Output$4.004 per 1M tokens
Cached input$0.3276 per 1M tokens

Example: 1,000 requests with 3,000 input and 1,000 output tokens each cost $7.83. Every rate is published and stays the same from one request to the next. See every model.

Rates as of 20 November 2026.

Specs

Made byZhipu AI
ReleasedAugust 2026
TypeChat
Context window1M tokens
Longest answer128K tokens
ReadsText
Thinks before answeringAlways on
WeightsOpen (GLM-5.3 License)
SizeAbout 744B total, 40B active
Processed inSingapore
Model idglm-5.3

Is GLM-5.3 worth the price?

For security analysis and command-line agents, it is worth testing. Z.ai reports 84.5% on the CyberGym security benchmark. Its headline coding gain is on a private benchmark whose method is not published, so try it on your own code.

For simple requests, look elsewhere. Every call reasons and there is no fast mode, so simple requests are slower and dearer than they need to be. It also reads text only.

GLM-5.3 and the models to compare it with, priced per 1M tokens
ModelInputOutputChoose it for
GLM-5.3$1.274$4.004Coding agents and security work.
DeepSeek-V4-Pro$1.56$3.12General coding, with a no-thinking mode for quick answers.
GLM-5.3-Flash$0.195$0.65A low-cost model that reads images and video.
Kimi K3$3.705$18.525Web research, images and video.

Prices per 1M tokens.

About GLM-5.3

GLM-5.3 is Z.ai’s top model, and an unusual kind of upgrade: the base model is the same as GLM-5.2, and every gain came from further training on top of it. The gains are largest where Z.ai aimed them: 66.9% on DeepSWE, up from 46.2%, and 84.5% on the CyberGym security benchmark, which Z.ai puts ahead of Claude Fable 5.

It always thinks before answering, at low, high or max effort (max is the default), and handles a million-token context. The weights are open under Z.ai’s own licence, which asks the very largest companies to pass a security review first.

What is GLM-5.3 good for?

  • Long coding agents that work through a large change.
  • Security analysis and defensive cyber work.
  • Command-line automation agents.

GLM-5.3 limitations

  • Every call reasons and there is no fast mode, so simple requests are slower and dearer than they need to be.
  • Z.ai’s headline coding gain is on a private benchmark whose method is not published.
  • Text only.

How to use the GLM-5.3 API

Use the OpenAI SDK you already have: point it at Roar AI and set the model name. Read the docs.

from openai import OpenAI

client = OpenAI(base_url="https://api.roar-ai.com/v1", api_key="roar_live_…")

reply = client.chat.completions.create(
    model="glm-5.3",
    messages=[{"role": "user", "content": "Summarise this contract in plain English."}],
)

GLM-5.3 FAQ

How much does GLM-5.3 cost?

Input is $1.274 per 1M tokens; output is $4.004 per 1M tokens. For example, 1,000 requests with 3,000 input and 1,000 output tokens each cost $7.83.

Can I use GLM-5.3 alongside other models?

Yes. Every model on Roar AI is on the same API key and the same monthly invoice, so switching from GLM-5.3 to another model only means changing the model name in your code.

Is GLM-5.3 open source?

Its weights are open under the GLM-5.3 License licence, so you could run it yourself. Through Roar AI you call it like any other model, with no servers to manage.

What is the context window of GLM-5.3?

GLM-5.3 can read up to 1M tokens in one request, and write an answer of up to 128K tokens.

Can GLM-5.3 read images and PDFs?

It reads text only. You can still send it a PDF: Roar AI turns each page into text first, at no extra charge. Scanned pages need a model that can see images.

Start using GLM-5.3

Create a free account, get an API key and call GLM-5.3 with the OpenAI SDK.