Instructions to use amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3 with Transformers:

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

pipe = pipeline("text-generation", model="amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3")
model = AutoModelForCausalLM.from_pretrained("amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3

SGLang

How to use amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3 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 "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3 with Docker Model Runner:
```
docker model run hf.co/amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3
```

Configuration Parsing Warning:In config.json: "quantization_config.bits" must be an integer

GLM-4.5-Air-Derestricted ExLlamaV3 Quantizations

ExLlamaV3 quantizations of GLM-4.5-Air-Derestricted with tensor-level (L3) optimization and boosted attention layers.

Using the provided measurement.json file and base quants, additional optimized quantizations can be made in seconds at any reasonable bpw. All work done with ExLlamaV3 v0.0.18.

Optimized

VRAM-targeted quants using exl3's measure.py → optimize.py → recompile.py pipeline with attention boost.

Branch	Size	bpw	Target
3.15bpw-h6-opt	41 GB	3.15	48GB @ 128k
4.37bpw-h6-opt	56 GB	4.37	64GB @ 128k
5.00bpw-h6-opt	64 GB	5.00	72GB @ 128k
6.33bpw-h6-opt	80 GB	6.33	96GB @ 128k

Note: The 6.33bpw quant hit the optimization ceiling - targeting 6.94bpw produced 6.18bpw pre-boost output (6.33 after attention boost), indicating all beneficial tensor swaps exhausted.

Base

Branch	Size	bpw
2.0bpw-h6	27 GB	2.0
3.0bpw-h6	39 GB	3.0
4.0bpw-h6	52 GB	4.0
5.0bpw-h6	64 GB	5.0
6.0bpw-h6	76 GB	6.0
7.0bpw-h6	88 GB	7.0

Downloads last month: 39

Model tree for amanwalksdownthestreet/GLM-4.5-Air-Derestricted-exl3

Base model

zai-org/GLM-4.5-Air

Finetuned

ArliAI/GLM-4.5-Air-Derestricted

Quantized

(25)

this model