Instructions to use metaXu264/generator-v2-prokaryote-3b-atlas-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use metaXu264/generator-v2-prokaryote-3b-atlas-ft with Transformers:

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

pipe = pipeline("text-generation", model="metaXu264/generator-v2-prokaryote-3b-atlas-ft")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("metaXu264/generator-v2-prokaryote-3b-atlas-ft")
model = AutoModelForMultimodalLM.from_pretrained("metaXu264/generator-v2-prokaryote-3b-atlas-ft")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use metaXu264/generator-v2-prokaryote-3b-atlas-ft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "metaXu264/generator-v2-prokaryote-3b-atlas-ft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "metaXu264/generator-v2-prokaryote-3b-atlas-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/metaXu264/generator-v2-prokaryote-3b-atlas-ft

SGLang

How to use metaXu264/generator-v2-prokaryote-3b-atlas-ft 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 "metaXu264/generator-v2-prokaryote-3b-atlas-ft" \
    --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": "metaXu264/generator-v2-prokaryote-3b-atlas-ft",
		"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 "metaXu264/generator-v2-prokaryote-3b-atlas-ft" \
        --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": "metaXu264/generator-v2-prokaryote-3b-atlas-ft",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use metaXu264/generator-v2-prokaryote-3b-atlas-ft with Docker Model Runner:
```
docker model run hf.co/metaXu264/generator-v2-prokaryote-3b-atlas-ft
```

GENERator-v2 Prokaryote 3B – Atlas Fine-tuned

This model is a fine-tuned version of:

GenerTeam/GENERator-v2-prokaryote-3b-base

Model Details

Finetuned from: GenerTeam/GENERator-v2-prokaryote-3b-base
Architecture: Decoder-only Transformer (Causal LM)
Tokenization: 6-mer DNA tokenizer (custom, requires trust_remote_code)
Domain: Prokaryotic genomic sequences
Fine-tuning steps: 36,000
Dataset: Atlas (custom)

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "<your-username>/generator-v2-prokaryote-3b-atlas-ft"

tokenizer = AutoTokenizer.from_pretrained(
    model_id,
    trust_remote_code=True
)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True
)

Citation

If you use the base model GENERator in your research, please cite the original paper:

@misc{wu2025generator,
  title        = {GENERator: A Long-Context Generative Genomic Foundation Model},
  author       = {Wei Wu and Qiuyi Li and Mingyang Li and Kun Fu and Fuli Feng and Jieping Ye and Hui Xiong and Zheng Wang},
  year         = {2025},
  eprint       = {2502.07272},
  archivePrefix= {arXiv},
  primaryClass = {cs.CL},
  url          = {https://arxiv.org/abs/2502.07272}
}

Downloads last month: 1

Paper for metaXu264/generator-v2-prokaryote-3b-atlas-ft

GENERator: A Long-Context Generative Genomic Foundation Model

Paper • 2502.07272 • Published Feb 11, 2025 • 4