GLM 4 32B

A cost-effective 32B foundation model with much stronger tool use, web search and code work than GLM 3. From the lab behind the THUDM models.

Jailbreak1 of 9levels resisted

Website OpenRouter HuggingFace Docs API GitHub

Provider: Zhipu AI
Released: 2025-07-24
Size: LARGE
Free API (OpenRouter): No
Per 1M tokens: $0.10 in · $0.10 out
Rival’s tab: $0.0059; 25/58 receipts survived +~$0.0094 from the pre-receipt era

Free API availability

No free endpoint for GLM 4 32B was found in Rival’s OpenRouter listings. Other providers or chat apps may have separate free offers.

Provider data: 13 Sep 2026OpenRouter usage limits

View API options

Get API accessProvider and language code samples

Provider

fromimport openai  OpenAI

client = OpenAI(
"https://openrouter.ai/api/v1"    base_url=,
"$OPENROUTER_API_KEY"    api_key=,
)

response = client.chat.completions.create(
"z-ai/glm-4-32b"    model=,
"role""user""content""Hello!"    messages=[{: , : }],
)
print(response.choices[0].message.content)

Set OPENROUTER_API_KEY with your OpenRouter API key from openrouter.ai/keys.

Deep analysisPsychometrics, taste index, writing DNA

SubjectiveBench

Taste Index

Across 58 scored outputs

How this is measured →

210.20xFloor

0100headroom →

Taste is judged on an uncapped scale, originality first. The space past 100 is craft today's models rarely reach.

Craft31

Originality16

Plays it safe

share of outputs that are the default answer

66%

Writing DNA

Stylometric Fingerprint

Based on 27 text responses

Tick = global average

Vocabulary Diversity55%

Unique words vs. total words. Higher = richer vocabulary.

Sentence Length14.7 words

Average words per sentence.

Hedging0.74

"Might", "perhaps", "arguably" per 100 words.

Bold Formatting4.9

**Bold** markers per 1,000 characters.

List Usage3.8

Bullet and numbered list items per 1,000 characters.

Section Structure0.35

Markdown headings per 1,000 characters.

Emoji Usage0.00

Emoji per 1,000 characters.

Transitions0.04

"However", "moreover", "furthermore" per 100 words.

Opening Habits

Consistency

78%

Across 27 responses

Favorites

Movie

The Shawshank Redemption

19948.7

Album

OK Computer

Radiohead

Book

Cien años de soledad

Gabriel García Márquez

City

Kyoto

Japan

Game

The Legend of Zelda: Ocarina of Time

Adventure, Action4.4

Model Responses

53 outputs · $0.0059 tracked across 25 receipts

Reviewing agent-written code?See a Brief PR report

Keep exploring

COMPARE

GLM 4 32B vs Ox Alpha

Real outputs compared side by side

RANKINGS

Best AI for Creative Writing

A stand-up set with puns banned, a manifesto against CAPTCHA, an interview wi...

Compare GLM 4 32B

Alternatives to GLM 4 32B

GLM 4 32B

Free API availability

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

GLM 4 32B vs Ox Alpha

Best AI for Creative Writing

Compare GLM 4 32B

Alternatives to GLM 4 32B

GLM 4 32B

Free API availability

Taste Index

Stylometric Fingerprint

Favorites

Model Responses

GLM 4 32B vs Ox Alpha

Best AI for Creative Writing