Instructions to use zai-org/GLM-Z1-Rumination-32B-0414 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zai-org/GLM-Z1-Rumination-32B-0414 with Transformers:

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

pipe = pipeline("text-generation", model="zai-org/GLM-Z1-Rumination-32B-0414")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zai-org/GLM-Z1-Rumination-32B-0414")
model = AutoModelForCausalLM.from_pretrained("zai-org/GLM-Z1-Rumination-32B-0414", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zai-org/GLM-Z1-Rumination-32B-0414 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zai-org/GLM-Z1-Rumination-32B-0414"
# 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-Z1-Rumination-32B-0414",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zai-org/GLM-Z1-Rumination-32B-0414

SGLang

How to use zai-org/GLM-Z1-Rumination-32B-0414 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-Z1-Rumination-32B-0414" \
    --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-Z1-Rumination-32B-0414",
		"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-Z1-Rumination-32B-0414" \
        --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-Z1-Rumination-32B-0414",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zai-org/GLM-Z1-Rumination-32B-0414 with Docker Model Runner:
```
docker model run hf.co/zai-org/GLM-Z1-Rumination-32B-0414
```

bug in demo code

by undefined-x - opened Apr 15, 2025

Discussion

undefined-x

Apr 15, 2025

three bug in demo code

def get_assistant():
    inputs = tokenizer.apply_chat_template(
        messages,
        return_tensors="pt",
        add_generation_prompt=True,
        return_dict=True,
    ).to(model.device)
    out = model.generate(input_ids=input["input_ids"], **generate_kwargs)
    return tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True).strip()

8th line shoud be out = model.generate(input_ids=inputs["input_ids"], **generate_kwargs) missing s in variable input

missing import re and import json... You use re module and json module in func get_func_name_args
should set "max_new_tokens" in generate_kwargs, because default prompt length(input+output) is 20 in method model.generate... It's useless in many scenarios. doc ref

Please review the code carefully before releasing the model.....

undefined-x

Apr 15, 2025

three bugs in func get_observation ,

. fucntion_name typo
. none return
missing enumerate in 7th line...

the correct func is following

def get_observation(function_name, args):
    if function_name == "search":
        mock_search_res = [
            {"title": "t1", "url":"url1", "snippet": "snippet_content_1"},
            {"title": "t2", "url":"url2", "snippet": "snippet_content_2"}
        ]
        content = "\n\n".join([f"【{i}†{res['title']}†{res['url']}\n{res['snippet']}】" for i, res in enumerate(mock_search_res)])
    elif function_name == "click":
        mock_click_res = "main content"
        content = mock_click_res
    elif function_name == "open":
        mock_open_res = "main_content"
        content = mock_open_res
    else:
        raise ValueError("unspport function name!")
        
    return content

please fix it.

ZHANGYUXUAN-zR

Z.ai org Apr 15, 2025

•

edited Apr 15, 2025

We have changed the code. Thank you for your support.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment