zai-org
/

GLM-5.2

Text Generation

Model card Files Files and versions

Instructions to use zai-org/GLM-5.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zai-org/GLM-5.2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="zai-org/GLM-5.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zai-org/GLM-5.2")
model = AutoModelForCausalLM.from_pretrained("zai-org/GLM-5.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use zai-org/GLM-5.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zai-org/GLM-5.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zai-org/GLM-5.2

How to use zai-org/GLM-5.2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "zai-org/GLM-5.2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "zai-org/GLM-5.2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zai-org/GLM-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zai-org/GLM-5.2 with Docker Model Runner:
```
docker model run hf.co/zai-org/GLM-5.2
```

Resources

View closed (8)

High Token Consumption and Cost Behavior in This Model

#33 opened 1 day ago by

Nice Model, can beat Gemini3.1pro,sonnet5 , and GPT-5.5 instant

#32 opened 1 day ago by

Dense 60-120B model

#31 opened 3 days ago by

Mya

#29 opened 4 days ago by

Multilingual model — testing for mobile deployment

#28 opened 4 days ago by

Will you provide a DFlash model?

#27 opened 4 days ago by

Deep admiration for your work—Any plans for a 30B-level GLM 5.2 Flash model for local edge execution?

#26 opened 6 days ago by

what is the purpose of this model

#25 opened 7 days ago by

redosempiternal66

Add ResearchClawBench evaluation result

#22 opened 9 days ago by

How to enable Turn-Level Thinking?

#21 opened 9 days ago by

glm-5.2-RedlineBench

#19 opened 10 days ago by

How much VRAM is needed to keep it running?

#18 opened 11 days ago by

GLM 5.2 Nano ( 4 to 20 B params) would be good for edge hardware

#16 opened 12 days ago by

GLM-5.2-Flash 20B -40B would be amazing.

#15 opened 13 days ago by

UniversalLove333

Is AIME 2026 Benchmark evaluated with Tool Call enabled?

#14 opened 14 days ago by

Multilingual or just English and Chinese

#13 opened 14 days ago by

How to reproduce the GLM-5.2 benchmarks across various datasets?

#11 opened 15 days ago by

crash

#9 opened 15 days ago by

Updating README.md ..

#8 opened 16 days ago by

Please FUCK Anthropic

#6 opened 16 days ago by

vision model

#5 opened 16 days ago by

This model feels good.

#4 opened 16 days ago by

We need some Air or at least some Flash

#3 opened 16 days ago by

Air or Flash model coming?

#2 opened 16 days ago by

Question regarding reasoning effort in chat template

#1 opened 16 days ago by