Instructions to use dotan1111/BetaInfer_Configuration1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dotan1111/BetaInfer_Configuration1 with Transformers:

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

pipe = pipeline("text-generation", model="dotan1111/BetaInfer_Configuration1")

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

tokenizer = AutoTokenizer.from_pretrained("dotan1111/BetaInfer_Configuration1")
model = AutoModelForCausalLM.from_pretrained("dotan1111/BetaInfer_Configuration1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dotan1111/BetaInfer_Configuration1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dotan1111/BetaInfer_Configuration1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dotan1111/BetaInfer_Configuration1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dotan1111/BetaInfer_Configuration1

SGLang

How to use dotan1111/BetaInfer_Configuration1 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 "dotan1111/BetaInfer_Configuration1" \
    --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": "dotan1111/BetaInfer_Configuration1",
		"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 "dotan1111/BetaInfer_Configuration1" \
        --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": "dotan1111/BetaInfer_Configuration1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dotan1111/BetaInfer_Configuration1 with Docker Model Runner:
```
docker model run hf.co/dotan1111/BetaInfer_Configuration1
```

dotan1111 commited on Aug 4

Commit

bc5a4b5

verified ·

1 Parent(s): 46a7dc1

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -17,7 +17,8 @@ tags:
 ## Abstract:
 Accurate inference of phylogenetic trees is fundamental to evolutionary biology, yet existing methods rely on complex pipelines involving multiple sequence alignment, explicit evolutionary models, and computationally intensive tree search procedures. Here, we present BetaInfer, a generative framework that reformulates phylogenetic tree inference as a sequence generation problem. BetaInfer leverages hybrid transformer-based architectures to directly map sets of unaligned sequences to phylogenetic trees represented in Newick format. Trained primarily on large-scale simulated evolutionary data with known ground truth, BetaInfer learns to capture complex evolutionary signals directly from sequence data. Ensemble-based generation of multiple candidate trees further improves robustness, reducing reconstruction error by over 30% relative to single predictions. Across extensive evaluations on both simulated and empirical datasets, BetaInfer achieves competitive performance relative to state-of-the-art phylogenetic pipelines, matching or exceeding the accuracy of established likelihood-based and distance-based methods under a wide range of conditions. Probing analyses further reveal that BetaInfer does not simply emulate classical distance-based algorithms, but instead encodes evolutionary relationships in a more integrated, global representation that supports direct tree generation. Together, these results demonstrate that generative models can serve as a viable and scalable alternative to standard phylogenetic pipelines.
-![alt text](https://github.com/technion-cs-nlp/BetaInfer/blob/main/outline_image.png)
 Illustration of predicting a phylogenetic tree using BetaInfer. Panel (a) shows the simulated “true” evolutionary dynamics, in which the ancestral sequence “AAMM” diverged along a phylogenetic tree evolving into four leaves: “ADAM”, “ADAW”, “AAM”, and “ATMMM”. Panel (b) illustrates the BetaReconstruct pipeline: (Ⅰ): unaligned protein sequences serve as input for our model; (Ⅱ): The unaligned protein sequences are concatenated into the input “sentence”, and the special token marking the start of the phylogenetic tree inference is added; (Ⅲ): Our trained model processes the input and generates the phylogenetic tree (the output); (Ⅳ): the generated phylogenetic tree. Panels (c) highlights the labeling of the taxa in the resulting tree, in which, the first protein sequence is labeled ℓ0, and the second is labeled ℓ1, and so forth.
 ## Public models

 ## Abstract:
 Accurate inference of phylogenetic trees is fundamental to evolutionary biology, yet existing methods rely on complex pipelines involving multiple sequence alignment, explicit evolutionary models, and computationally intensive tree search procedures. Here, we present BetaInfer, a generative framework that reformulates phylogenetic tree inference as a sequence generation problem. BetaInfer leverages hybrid transformer-based architectures to directly map sets of unaligned sequences to phylogenetic trees represented in Newick format. Trained primarily on large-scale simulated evolutionary data with known ground truth, BetaInfer learns to capture complex evolutionary signals directly from sequence data. Ensemble-based generation of multiple candidate trees further improves robustness, reducing reconstruction error by over 30% relative to single predictions. Across extensive evaluations on both simulated and empirical datasets, BetaInfer achieves competitive performance relative to state-of-the-art phylogenetic pipelines, matching or exceeding the accuracy of established likelihood-based and distance-based methods under a wide range of conditions. Probing analyses further reveal that BetaInfer does not simply emulate classical distance-based algorithms, but instead encodes evolutionary relationships in a more integrated, global representation that supports direct tree generation. Together, these results demonstrate that generative models can serve as a viable and scalable alternative to standard phylogenetic pipelines.
+![Figure1](https://cdn-uploads.huggingface.co/production/uploads/63047e2d412a1b9d381b045d/k-X_8QTZY3j-1oMVM5f-P.png)
 Illustration of predicting a phylogenetic tree using BetaInfer. Panel (a) shows the simulated “true” evolutionary dynamics, in which the ancestral sequence “AAMM” diverged along a phylogenetic tree evolving into four leaves: “ADAM”, “ADAW”, “AAM”, and “ATMMM”. Panel (b) illustrates the BetaReconstruct pipeline: (Ⅰ): unaligned protein sequences serve as input for our model; (Ⅱ): The unaligned protein sequences are concatenated into the input “sentence”, and the special token marking the start of the phylogenetic tree inference is added; (Ⅲ): Our trained model processes the input and generates the phylogenetic tree (the output); (Ⅳ): the generated phylogenetic tree. Panels (c) highlights the labeling of the taxa in the resulting tree, in which, the first protein sequence is labeled ℓ0, and the second is labeled ℓ1, and so forth.
 ## Public models