Instructions to use GenerTeam/GENERator-v2-eukaryote-3b-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GenerTeam/GENERator-v2-eukaryote-3b-base with Transformers:

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

pipe = pipeline("text-generation", model="GenerTeam/GENERator-v2-eukaryote-3b-base", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("GenerTeam/GENERator-v2-eukaryote-3b-base", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("GenerTeam/GENERator-v2-eukaryote-3b-base", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use GenerTeam/GENERator-v2-eukaryote-3b-base with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/GenerTeam/GENERator-v2-eukaryote-3b-base

SGLang

How to use GenerTeam/GENERator-v2-eukaryote-3b-base 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 "GenerTeam/GENERator-v2-eukaryote-3b-base" \
    --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": "GenerTeam/GENERator-v2-eukaryote-3b-base",
		"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 "GenerTeam/GENERator-v2-eukaryote-3b-base" \
        --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": "GenerTeam/GENERator-v2-eukaryote-3b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use GenerTeam/GENERator-v2-eukaryote-3b-base with Docker Model Runner:
```
docker model run hf.co/GenerTeam/GENERator-v2-eukaryote-3b-base
```

GenerTeam commited on Feb 2

Commit

b875aec

verified ·

1 Parent(s): b0b9198

Update README.md

Browse files

Files changed (1) hide show

README.md +18 -9

README.md CHANGED Viewed

@@ -21,7 +21,7 @@ We apologize for any inconvenience this may cause and recommend adhering to the
 ## Abouts
-In this repository, we present GENERator-v2, a generative genomic foundation with enhanced performance in eukaryotic domain. More technical details are coming soon...
 Python scripts for downstream analysis are available on Github: [https://github.com/GenerTeam/GENERator](https://github.com/GenerTeam/GENERator).
@@ -162,13 +162,22 @@ print("Mean Pooling Embeddings:", mean_embeddings)
 ## Citation
 ```
-@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},
 }
 ```

 ## Abouts
+In this repository, we present GENERator-v2, a generative genomic foundation with enhanced performance in eukaryotic domain. More technical details are provided in the GENERator-v2 [technical report](https://www.biorxiv.org/content/10.64898/2026.01.27.702015v1).
 Python scripts for downstream analysis are available on Github: [https://github.com/GenerTeam/GENERator](https://github.com/GenerTeam/GENERator).
 ## Citation
 ```
+@article {li2026generator2,
+	author = {Li, Qiuyi and Zhan, Zhihao and Feng, Shikun and Zhu, Yiheng and He, Yuan and Wu, Wei and Shi, Zhenghang and Wang, Shengjie and Hu, Zongyong and Yang, Zhao and Li, Jiaoyang and Tang, Jian and Liu, Haiguang and Qin, Tao},
+	title = {Functional In-Context Learning in Genomic Language Models with Nucleotide-Level Supervision and Genome Compression},
+	elocation-id = {2026.01.27.702015},
+	year = {2026},
+	doi = {10.64898/2026.01.27.702015},
+	publisher = {Cold Spring Harbor Laboratory},
+	URL = {https://www.biorxiv.org/content/early/2026/01/29/2026.01.27.702015},
+	journal = {bioRxiv}
 }
+@article{wu2025generator,
+  title={GENERator: a long-context generative genomic foundation model},
+  author={Wu, Wei and Li, Qiuyi and Li, Mingyang and Fu, Kun and Feng, Fuli and Ye, Jieping and Xiong, Hui and Wang, Zheng},
+  journal={arXiv preprint arXiv:2502.07272},
+  year={2025}
+}
 ```