Instructions to use defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix with Transformers:

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

pipe = pipeline("text-generation", model="defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix")
model = AutoModelForCausalLM.from_pretrained("defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix", 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 defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "defiobi00/GLM-5.3-DERISKED-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": "defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix

SGLang

How to use defiobi00/GLM-5.3-DERISKED-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 "defiobi00/GLM-5.3-DERISKED-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": "defiobi00/GLM-5.3-DERISKED-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 "defiobi00/GLM-5.3-DERISKED-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": "defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix with Docker Model Runner:
```
docker model run hf.co/defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix
```

Access request

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

This repository is public metadata with a manual download gate. Submit an accurate intended-use statement. Approval is discretionary and may be revoked for misuse or redistribution without authorization.

GLM-5.3-DERISKED-Int4-Int8Mix

Client drop · full GLM-5.3 MoE · compressed-tensors pack-quant · Blackfrost DWM a=3

Built by Blackfrost · Las Vegas, Nevada

Public metadata, manual download gate. Client delivery. Request access on the Hub. Do not redistribute the weights without authorization.

What this is

Independent Int4-Int8Mix pack-quantized safetensors of GLM-5.3 (full ~753B MoE, not Flash) after a Blackfrost direction-weight modification (DWM) pass. Source mix weights were not overwritten.

The intended behavior is in the weights. Production DWM details are proprietary and are not disclosed beyond the locked recipe below.

This artifact has not been through a judged refusal suite. Do not copy NVFP4/BF16/Q3 refusal percentages onto this checkpoint.

Specifications


Architecture	`GlmMoeDsaForCausalLM` (`glm_moe_dsa`)
Quant	`compressed-tensors` `pack-quantized` · int8 group-128 linears · int4 group-128 routed experts · layer 0 BF16 · MTP int8 channel (stock)
Shards	282 safetensors
Layers	78 main + MTP layer 78
Hidden size	6144
Experts	256 routed · top-8 / token · 1 shared
DWM	alpha 3.0 · skip-early 2 · one pass · frozen scales · no norm restore
Targets	76 `o_proj` + 1 dense `down_proj` + 19200 expert `down_proj` + 75 shared `down_proj`
Runtime	vLLM compressed-tensors / pack-quantized (multi-Spark). Not a GGUF.
Languages	English and Chinese

Sibling client GGUF (separate repo, do not mix formats): defiobi00/GLM-5.3-DERISKED-UD-Q3_K_XL.

Downloads last month: 3

Safetensors

Model size

779B params

Tensor type

I32

BF16

F32

Model tree for defiobi00/GLM-5.3-DERISKED-Int4-Int8Mix

Base model

zai-org/GLM-5.3-BF16

Quantized

(23)

this model