ModelSheep
All models

GLM 5.3 API

Z.ai's open-weights model for coding agents and long tool-use tasks, called from the OpenAI SDK, from $0.915 per million input tokens.

Get API keyRead the docs
  • 1.3M token context
  • 753.3B total parameters
  • Served in FP8
  • GLM 5.3 licence
from openai import OpenAI

client = OpenAI(
    base_url="https://api.modelsheep.com",
    api_key=API_KEY_MODELSHEEP,
)

response = client.chat.completions.create(
    model="zai-org/GLM-5.3",
    messages=[
        {
            "role": "user",
            "content": "Why do sheep count people?"
        }
    ],
)

print(response.choices[0].message.content)

How to use the GLM 5.3 API

  1. 1

    Create a ModelSheep account and copy an API key from the dashboard.

  2. 2

    Point the OpenAI SDK at the base url https://api.modelsheep.com and set the model to zai-org/GLM-5.3.

  3. 3

    Send your messages to /chat/completions with reasoning_effort set, then read the reply from the first choice.

What GLM 5.3 is

GLM 5.3 is an open-weights mixture of experts language model from Z.ai. It uses the same base model as GLM 5.2, and Z.ai states that every gain comes from post-training. Z.ai built it for coding and for long-horizon agentic work, where one task runs many tool calls in sequence. It is the large model in the family and it takes text only, while GLM 5.3 Flash is smaller and also reads images and video.

What it does well

  • Drives a terminal agent through a multi-step task. It runs commands and
  • Fixes a bug across a repository from an issue description alone.
  • Finds a memory safety bug in a C codebase and writes a working exploit.

What it is not for

  • Image, audio or video input. GLM 5.3 takes text only.
  • Short, latency-sensitive calls. Thinking cannot be switched off. Every

Calling it

Call GLM 5.3 on POST /chat/completions with the model string zai-org/GLM-5.3. The parameter callers miss is reasoning_effort, which takes low, high or max and defaults to max, so a request that omits it spends the largest thinking budget. Thinking cannot be switched off on this model, so send low when you want a short reply at a lower cost.

zai-org/GLM-5.3 /chat/completions

SDKs and languages

GLM 5.3 answers on an HTTP endpoint, so any language that can post JSON reaches it. These are the ones the box above writes out.

OpenAI SDK
PythonTypeScriptGoJavaC#PHPRust
Anthropic SDK
PythonTypeScriptGoJavaC#PHPRust
HTTP
cURLPythonTypeScriptGoRustJavaC#PHP

Specifications

AuthorZ.ai
Categorylanguage
Acceptstext
Returnstext
Total parameters753.3B
Active parametersdense, all of them
Score44.9 on Artificial Analysis intelligence index
LanguagesEnglish, ChineseRead off the model card metadata. Z.ai writes no sentence about language coverage.
Precision servedFP8
Context1.3M tokens
Vectorn/a
LicenseGLM 5.3

Price

Input

$0.915

per million tokens

Output

$2.875

per million tokens

Against OpenRouter

10%

below the cheapest of 26 providers

The cheapest OpenRouter price for this model was $1.017 in and $3.195 out, with a median of $1.400 and $4.400 across the comparable providers. Read on 2026-09-17.

License

Z.ai ships GLM 5.3 under its own licence, GLM 5.3, not a standard open one. It permits use, modification, distribution and commercial use, but an operator of a model as a service business above 10 billion US dollars of revenue in any 12 months must first pass a Z.ai security review, so read the licence file before you ship.

Similar models

Kimi K3moonshotai/Kimi-K3
Parameters
2.8T
Context
1M
Input per 1M
$2.250
Output per 1M
$11.475
License
Kimi K3
GLM 5.3 Flashzai-org/GLM-5.3-Flash
Parameters
321.3B, 18B
Context
1.3M
Input per 1M
$0.081
Output per 1M
$0.270
License
MIT
Qwen3.8 2.4T A95BQwen/Qwen3.8-2.4T-A95B
Parameters
2.4T, 95B
Context
1M
Input per 1M
$1.800
Output per 1M
$5.400
License
Qwen3.8 Max

Start using GLM 5.3

Point the OpenAI SDK at api.modelsheep.com. Your prompts stay in Germany, and nothing is retained.

Get API keyRead the quickstart