Instructions to use lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated with Transformers:

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

pipe = pipeline("image-text-to-text", model="lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated")
model = AutoModelForMultimodalLM.from_pretrained("lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated

SGLang

How to use lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated 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 "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated" \
    --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": "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated" \
        --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": "lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated with Docker Model Runner:
```
docker model run hf.co/lovesenko/GLM-5.3-Flash-tr3-4bpw-Abliterated
```

GLM-5.3-Flash-tr3-4bpw-Abliterated / PROVENANCE.md

lovesenko

Add files using upload-large-folder tool

ca44eaa verified 2 days ago

preview code

Raw

History Blame

1.34 kB

Provenance and attribution

This release uses transparent, content-addressed provenance. It does not use a hidden watermark, phone-home request, telemetry, or inference-output marker.

The current v84 image embeds /opt/glm53/PROVENANCE.json and carries standard OCI source, revision, author, documentation, version, checkpoint, and release labels. The manifest binds the corrected Triton DFlash mask, INT4 attention call path, vision RoPE fallback, multimodal template, and validation receipts with SHA-256 hashes. The registry digest binds the complete published image.

Inspect and verify a pulled image with:

runtime/verify-provenance.sh \
  verdictai/glm53-flash-exl3-k4:r19-sm120-tp2-ep2-dcp2-v84-dflash2@sha256:0f1cdcc8891f1cc3a444121eb61d366289a1cbba285f0892dcbb24bc94961692

The canonical public history is the combination of:

the immutable Docker/OCI digest and attached build attestations;
the Git commit history and release receipt in this repository;
the Hugging Face model revision and copied release receipt;
the hashes inside the embedded provenance manifest.

These records make copied or modified artifacts correlatable; they do not make removal impossible and do not, by themselves, prove legal infringement. The applicable attribution and provenance-retention requirements are in LICENSE.