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
```

JunHowie commited on 6 days ago

Commit

8e7a305

verified ·

1 Parent(s): 8677dbb

Upload README.md

Browse files

Files changed (1) hide show

README.md +8 -6

README.md CHANGED Viewed

@@ -39,10 +39,11 @@ The default `reasoning_effort` is changed to `medium-high` to reduce thinking-to
 Accuracy reference, using the [SGLang GLM-5.2 FP8 H200 / default / low-latency / single-node AIME25 recipe](https://lmsysorg.mintlify.app/cookbook/autoregressive/GLM/GLM-5.2#hw=h200&variant=default&quant=fp8&strategy=low-latency&nodes=single):
-| Model | Quantization | Reasoning effort | AIME25 pass@1 |
-|-------|--------------|------------------|---------------|
-| ZhipuAI/GLM-5.2-FP8 | FP8 | `max` | `87.7%` |
-| tclf90/GLM-5.2-Int4-Int8Mix | Int4-Int8Mix, W4A16/W8A16 | `medium-high` | `86.46%` |
 This is a lightweight reproduction reference rather than a full formal benchmark.
@@ -50,15 +51,16 @@ This is a lightweight reproduction reference rather than a full formal benchmark
 ```
 vllm==0.23.0
 ```
-As of **2026-06-21**, this model has been verified on an 8 x H200 machine with a Python 3.12 virtual environment and vLLM 0.23.0.
 Create a fresh Python environment and install vLLM:
 ```
 python3.12 -m venv venv
 source venv/bin/activate
-pip install vllm==0.23.0
 ```
 [vLLM Official Guide](https://recipes.vllm.ai/zai-org/GLM-5.2)

 Accuracy reference, using the [SGLang GLM-5.2 FP8 H200 / default / low-latency / single-node AIME25 recipe](https://lmsysorg.mintlify.app/cookbook/autoregressive/GLM/GLM-5.2#hw=h200&variant=default&quant=fp8&strategy=low-latency&nodes=single):
+| Model | Runtime | Quantization | Reasoning effort | AIME25 pass@1 |
+|-------|---------|--------------|------------------|---------------|
+| ZhipuAI/GLM-5.2-FP8 | SGLang | FP8 | `max` | `87.7%` |
+| tclf90/GLM-5.2-Int4-Int8Mix | vLLM | Int4-Int8Mix, W4A16/W8A16 | `max` | `92.92%` |
+| tclf90/GLM-5.2-Int4-Int8Mix | vLLM | Int4-Int8Mix, W4A16/W8A16 | `medium-high` | `86.46%` |
 This is a lightweight reproduction reference rather than a full formal benchmark.
 ```
 vllm==0.23.0
+transformers==5.12.1
 ```
+As of **2026-06-21**, this model has been verified on an 8 x H200 machine with a Python 3.12 virtual environment, vLLM 0.23.0, and Transformers 5.12.1.
 Create a fresh Python environment and install vLLM:
 ```
 python3.12 -m venv venv
 source venv/bin/activate
+pip install vllm==0.23.0 transformers==5.12.1
 ```
 [vLLM Official Guide](https://recipes.vllm.ai/zai-org/GLM-5.2)