Instructions to use skypro1111/m2m100-ukr-verbalization-ct2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use skypro1111/m2m100-ukr-verbalization-ct2 with Transformers:

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

pipe = pipeline("text-generation", model="skypro1111/m2m100-ukr-verbalization-ct2")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("skypro1111/m2m100-ukr-verbalization-ct2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use skypro1111/m2m100-ukr-verbalization-ct2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "skypro1111/m2m100-ukr-verbalization-ct2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "skypro1111/m2m100-ukr-verbalization-ct2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/skypro1111/m2m100-ukr-verbalization-ct2

SGLang

How to use skypro1111/m2m100-ukr-verbalization-ct2 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 "skypro1111/m2m100-ukr-verbalization-ct2" \
    --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": "skypro1111/m2m100-ukr-verbalization-ct2",
		"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 "skypro1111/m2m100-ukr-verbalization-ct2" \
        --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": "skypro1111/m2m100-ukr-verbalization-ct2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use skypro1111/m2m100-ukr-verbalization-ct2 with Docker Model Runner:
```
docker model run hf.co/skypro1111/m2m100-ukr-verbalization-ct2
```

skypro1111 commited on Apr 30, 2025

Commit

ff98eca

verified ·

1 Parent(s): c5e6b71

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -9

README.md CHANGED Viewed

@@ -69,7 +69,7 @@ text = "Моя бабуся народилася 07.11.1919, у важкий п
 result = process_sentence(translator, tokenizer, text)
 print(result)
 # Output: "Моя бабуся народилася сьомого листопада тисяча дев'ятсот дев'ятнадцятого року, у важкий післявоєнний час."
 ## Performance Benchmark
 ### Original Model (RTX 3090Ti, FP16)
@@ -170,14 +170,6 @@ Total time: 1.80 seconds
 Average time per sentence: 0.164 seconds
 ```
-### Performance Comparison
-| Metric | Original Model | CTranslate2 Model | Speedup |
-|--------|---------------|-------------------|---------|
-| Average time per sentence | 0.590 seconds | 0.164 seconds | 3.6x |
-| Total time (10 sentences) | 5.90 seconds | 1.80 seconds | 3.3x |
-| Memory usage (estimated) | ~2.5GB | ~800MB | 3.1x |
 The CTranslate2 optimization provides significant performance improvements while maintaining the same quality of verbalization.
 ## Model Information

 result = process_sentence(translator, tokenizer, text)
 print(result)
 # Output: "Моя бабуся народилася сьомого листопада тисяча дев'ятсот дев'ятнадцятого року, у важкий післявоєнний час."
+```
 ## Performance Benchmark
 ### Original Model (RTX 3090Ti, FP16)
 Average time per sentence: 0.164 seconds
 ```
 The CTranslate2 optimization provides significant performance improvements while maintaining the same quality of verbalization.
 ## Model Information