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

Libraries

How to use Hailay/GeezScriptTokenizer with Transformers:

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

pipe = pipeline("text-generation", model="Hailay/GeezScriptTokenizer")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Hailay/GeezScriptTokenizer")
model = AutoModelForMultimodalLM.from_pretrained("Hailay/GeezScriptTokenizer")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Hailay/GeezScriptTokenizer with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Hailay/GeezScriptTokenizer

SGLang

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

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

Hailay commited on May 17, 2025

Commit

b6ae2e9

verified ·

1 Parent(s): f2ecb69

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -9

README.md CHANGED Viewed

@@ -9,12 +9,4 @@ datasets:
 ---
 **1.Model Description**
-**Hailay/GeezScriptTokenizer**: is a language-specific tokenizer developed to handle the unique characteristics of Geez script languages, particularly Amharic and Tigrinya. This tokenizer is designed to effectively manage the complexities of these languages by accurately identifying and processing prefixes, postfixes, and word boundaries within the text. By incorporating these language-specific rules, GeezScriptTokenizer significantly improves tokenization efficiency, ensuring better representation and performance for tasks involving Amharic and Tigrinya.
-This tokenizer is highly suited for natural language processing (NLP) tasks where standard multilingual tokenizers may struggle with the nuances of Geez script languages. Hailay/GeezScriptTokenizer is an ideal tool for researchers and developers working with these languages, providing a tailored approach to tokenization that enhances the overall quality of language models and downstream tasks.
-##2. How To Use #
-To use Hailay/GeezScriptTokenizer, you can load it from Hugging Face’s Transformers library with just a few lines of code:
-Don't forget to fix the encoding method.

 ---
 **1.Model Description**
+**Hailay/GeezScriptTokenizer**: is a language-specific tokenizer developed to handle the unique characteristics of Geez script languages.