Instructions to use LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2 with Transformers:

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

pipe = pipeline("text-generation", model="LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2")
model = AutoModelForCausalLM.from_pretrained("LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2", 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 LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2

SGLang

How to use LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2 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 "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2" \
    --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": "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2",
		"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 "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2" \
        --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": "LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2 with Docker Model Runner:
```
docker model run hf.co/LatentWanderer/THUDM_GLM-4-32B-0414-6.5bpw-h8-exl2
```

Gibberish

by gtkunit - opened Apr 23, 2025

Discussion

gtkunit

Apr 23, 2025

Hi, thanks for the quant. I've tried to convert it myself but got gibberish. The same thing happens for me with your quant (exl3 works fine).
I'm on the latest exl2 dev branch. TabbyAPI or examples/chat.py give the same result.
Did you get it to work?

LatentWanderer

Owner Apr 23, 2025

Hi, im very sorry, i didn't try the model before, only the exl3. The output is for me gibberish as well, there is now an issue for that.

LatentWanderer

Owner Apr 23, 2025

Looks like exl3 is the way to go: https://github.com/turboderp-org/exllamav2/issues/772#issuecomment-2825374088

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