Instructions to use QuantTrio/GLM-5.2-Int4-Int8Mix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantTrio/GLM-5.2-Int4-Int8Mix with Transformers:

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

pipe = pipeline("text-generation", model="QuantTrio/GLM-5.2-Int4-Int8Mix")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("QuantTrio/GLM-5.2-Int4-Int8Mix")
model = AutoModelForCausalLM.from_pretrained("QuantTrio/GLM-5.2-Int4-Int8Mix")
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 QuantTrio/GLM-5.2-Int4-Int8Mix with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantTrio/GLM-5.2-Int4-Int8Mix"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantTrio/GLM-5.2-Int4-Int8Mix",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantTrio/GLM-5.2-Int4-Int8Mix

SGLang

How to use QuantTrio/GLM-5.2-Int4-Int8Mix 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 "QuantTrio/GLM-5.2-Int4-Int8Mix" \
    --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": "QuantTrio/GLM-5.2-Int4-Int8Mix",
		"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 "QuantTrio/GLM-5.2-Int4-Int8Mix" \
        --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": "QuantTrio/GLM-5.2-Int4-Int8Mix",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QuantTrio/GLM-5.2-Int4-Int8Mix with Docker Model Runner:
```
docker model run hf.co/QuantTrio/GLM-5.2-Int4-Int8Mix
```

when apply the mtp parameters, and the accuracy of aime25 and gpqa-diamond will decline

by youngmd - opened 4 days ago

Discussion

youngmd

4 days ago

using the parameters in A100*8 from the link https://huggingface.co/QuantTrio/GLM-5.2-Int4-Int8Mix/discussions/1, also find the result of some codes is not good when apply the mtp parameters.

aime25 decline from 0.93 to 0.67
gpqa-dimaond decline from 0.92 to 0.79

        --served-model-name GLM-5.2
        --dtype bfloat16
        --quantization compressed-tensors
        # --kv-cache-dtype fp8 # Triton Sparse MLA do not support fp8
        --tensor-parallel-size 8
        --enable-expert-parallel
        --max-model-len auto
        --gpu-memory-utilization 0.96
        --max-num-seqs 4
        --tool-call-parser glm47
        --reasoning-parser glm45
         --speculative-config '{"method":"mtp","num_speculative_tokens":1}' # do not enable Speculative Decoding, it will reduce the context length to 100k.
        --disable-uvicorn-access-log
        --safetensors-load-strategy prefetch

youngmd

4 days ago

when apply the mtp, the avg output will increase from 39 tok/s to 60 tok/s, however the accuracy will decline.

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