Instructions to use zai-org/GLM-5.1-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zai-org/GLM-5.1-FP8 with Transformers:

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

pipe = pipeline("text-generation", model="zai-org/GLM-5.1-FP8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zai-org/GLM-5.1-FP8")
model = AutoModelForCausalLM.from_pretrained("zai-org/GLM-5.1-FP8")
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

vLLM

How to use zai-org/GLM-5.1-FP8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zai-org/GLM-5.1-FP8"
# 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-5.1-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zai-org/GLM-5.1-FP8

SGLang

How to use zai-org/GLM-5.1-FP8 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-5.1-FP8" \
    --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-5.1-FP8",
		"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-5.1-FP8" \
        --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-5.1-FP8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zai-org/GLM-5.1-FP8 with Docker Model Runner:
```
docker model run hf.co/zai-org/GLM-5.1-FP8
```

Update chat_template.jinja

by quark2world - opened Apr 10

base: refs/heads/main

←

from: refs/pr/3

Discussion Files changed

+10

-0

quark2world

Apr 10

vLLM converts tool message content from string to list of content parts
[{"type": "text", "text": "..."}] before passing to the chat template.

The template only checks m.content is string, so tool results fall
into the else branch and render as empty . The model
then cannot see any tool output.

This adds handling for the iterable content parts format between the
existing string and else branches.

Affects vLLM tool calling with GLM-5.1 and likely GLM-5 as well.

Update chat_template.jinja6ad52ee1

ZHANGYUXUAN-zR

Z.ai org Apr 16

Thanks for identifying this issue and taking the time to submit a fix!

Unfortunately we can't merge this PR as-is, because the new elif m.content is iterable and m.content is not mapping branch is too broad — it also captures tool_reference content (used by our defer_loading feature), which causes that path to silently produce an empty <tool_response></tool_response> and breaks deferred tool loading entirely.

We've already pushed an updated template that addresses the array content handling while preserving the tool_reference path. Thanks again for the report and the contribution — really appreciated!

ZHANGYUXUAN-zR changed pull request status to closed Apr 16

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