zai-org
/

GLM-4.7

Text Generation

Model card Files Files and versions

Instructions to use zai-org/GLM-4.7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zai-org/GLM-4.7 with Transformers:

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

pipe = pipeline("text-generation", model="zai-org/GLM-4.7")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zai-org/GLM-4.7")
model = AutoModelForCausalLM.from_pretrained("zai-org/GLM-4.7")
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

How to use zai-org/GLM-4.7 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zai-org/GLM-4.7"
# 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-4.7",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/zai-org/GLM-4.7

How to use zai-org/GLM-4.7 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-4.7" \
    --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-4.7",
		"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-4.7" \
        --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-4.7",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use zai-org/GLM-4.7 with Docker Model Runner:
```
docker model run hf.co/zai-org/GLM-4.7
```

Resources

View closed (13)

Create .eval_results/apex-agents.yaml

#53 opened 28 days ago by

Create .eval_results/apex-v1-extended.yaml

#52 opened 29 days ago by

Add YC-Bench benchmark result (avg $398,410)

#51 opened about 2 months ago by

Add SWE-bench Verified evaluation result (73.8%)

#50 opened 2 months ago by

Create apex-agents.yaml

#48 opened 2 months ago by

Install & run zai-org/GLM-4.7 easily using llmpm

#47 opened 2 months ago by

fix: `clear_thinking` inserts spurious `</think>` tag when reasoning_content is empty

#46 opened 2 months ago by

Add Terminal-Bench evaluation result (33.4%)

#44 opened 3 months ago by

Create .eval_results/evasionbench.yaml

#43 opened 3 months ago by

如果GLM也出一个80B的开源模型，下载量应该也会在一周迅速突破十万。

#42 opened 3 months ago by

Does this model support MLA or only the flash version does?

#41 opened 3 months ago by

Add HLE evaluation result (with tools)

#40 opened 4 months ago by

Use message.reasoning when found

#39 opened 4 months ago by

Add MMLU-Pro evaluation result

#38 opened 4 months ago by

which is the context length of GLM-4.7 ? 202752 or 128000 ?

#33 opened 4 months ago by

本地部署开源版本GLM-4.7并接入claude命令使用，就差最后一步！！！

#32 opened 4 months ago by

When running GLM-4.7 with sglang-0.5.6 to process a /v1/chat/completions request, it encounters a BadRequestError

#30 opened 4 months ago by

Base Model / Air

#28 opened 4 months ago by

# integration status unknown. # Please clone model and use locally. # Also feel free to open a Pull request # for integration of the huggingface model hub # into the corresponding library =)

#26 opened 4 months ago by

使用这个模型编写代码的视频，如果你也想用它来编程的话

#25 opened 5 months ago by

Is Z.AI slow as molasses ? Do not buy yearly subsciption!!!

#24 opened 5 months ago by

New tool to benchmark GGUF files

#23 opened 5 months ago by

Rename README.md to O1.md

#21 opened 5 months ago by

MLX[LMStudio] Enable Preserved Thinking Mode ?

#20 opened 5 months ago by

Unable to produce correct `object` type tool call param

#18 opened 5 months ago by

Production deployment considerations

#17 opened 5 months ago by

Request Details on GPU ad memory requirements

#16 opened 5 months ago by

As a coding assistant in VS Code, it was disturbing.

#15 opened 5 months ago by

Jinja模板是不是有问题

#14 opened 5 months ago by

LM-4.7测评视频

#12 opened 5 months ago by

best coding&reasoning model. thank you Z.AI

#10 opened 5 months ago by

Browsecomp/HLE Reproducibility | 结果复现

#9 opened 5 months ago by

This model is Awesome, it's has get a hug progress in coding...

#8 opened 5 months ago by

The scores for K2-thinking and MiMo-V2-Flash were mixed up

#7 opened 5 months ago by

I'm even more looking forward to the GLM4.7 Air.

#6 opened 5 months ago by

Can we get a version that doesn't have the propaganda ?

#5 opened 5 months ago by

GGUF support is here!

#4 opened 5 months ago by

Video of Step-by-Step Review and Testing

#3 opened 5 months ago by

Is the GLM series a coding only model now?

#2 opened 5 months ago by

Doctor-Chad-PhD

we still need some Air

#1 opened 5 months ago by