Instructions to use fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym with Transformers:

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

pipe = pipeline("text-generation", model="fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym")
model = AutoModelForMultimodalLM.from_pretrained("fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym

SGLang

How to use fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym 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 "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym" \
    --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": "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym",
		"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 "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym" \
        --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": "fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym with Docker Model Runner:
```
docker model run hf.co/fbaldassarri/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym
```

tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym / README.md

fbaldassarri

Initial Upload

109e70c verified over 1 year ago

preview code

Raw

History Blame Contribute Delete

2.32 kB

	---
	license: other
	license_name: falcon-llm-license
	license_link: https://falconllm.tii.ae/falcon-terms-and-conditions.html
	language:
	- en
	- fr
	- es
	- pt
	pipeline_tag: text-generation
	tags:
	- causal-lm
	- autoround
	- auto-round
	- intel-autoround
	- woq
	- intel
	- pytorch
	- falcon3
	model_name: Falcon3 7B Base
	base_model:
	- tiiuae/Falcon3-7B-Base
	inference: false
	library_name: transformers
	model_creator: tiiuae
	prompt_template: '{prompt} '
	quantized_by: fbaldassarri
	---

	## Model Information

	Quantized version of [tiiuae/Falcon3-7B-Base](https://huggingface.co/tiiuae/Falcon3-7B-Base) using torch.float32 for quantization tuning.
	- 4 bits (INT4)
	- group size = 128
	- Asymmetrical Quantization
	- Method WoQ (AutoRound format)

	Fast and low memory, 2-3X speedup (slight accuracy drop at W4G128)

	Quantization framework: [Intel AutoRound](https://github.com/intel/auto-round) v0.4.4

	Note: this INT4 version of Falcon3-7B-Base has been quantized to run inference through CPU.

	## Replication Recipe

	### Step 1 Install Requirements

	I suggest to install requirements into a dedicated python-virtualenv or a conda enviroment.

	```
	wget https://github.com/intel/auto-round/archive/refs/tags/v0.4.4.tar.gz
	tar -xvzf v0.4.4.tar.gz
	cd auto-round-0.4.4
	pip install -r requirements-cpu.txt --upgrade
	```

	### Step 2 Build Intel AutoRound wheel from sources

	```
	pip install -vvv --no-build-isolation -e .[cpu]
	```

	### Step 3 Script for Quantization

	```
	from transformers import AutoModelForCausalLM, AutoTokenizer
	model_name = "tiiuae/Falcon3-7B-Base"
	model = AutoModelForCausalLM.from_pretrained(model_name)
	tokenizer = AutoTokenizer.from_pretrained(model_name)
	from auto_round import AutoRound
	bits, group_size, sym, device, amp = 4, 128, False, 'cpu', False
	autoround = AutoRound(model, tokenizer, nsamples=128, iters=200, seqlen=512, batch_size=4, bits=bits, group_size=group_size, sym=sym, device=device, amp=amp)
	autoround.quantize()
	output_dir = "./AutoRound/tiiuae_Falcon3-7B-Base-autoround-int4-gs128-asym"
	autoround.save_quantized(output_dir, format='auto_round', inplace=True)
	```

	## License

	[Falcon3 License](https://falconllm.tii.ae/falcon-terms-and-conditions.html)

	## Disclaimer

	This quantized model comes with no warranty. It has been developed only for research purposes.