Instructions to use laion/anh-bloomz-7b1-mt-cross-lingual with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use laion/anh-bloomz-7b1-mt-cross-lingual with Transformers:

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

pipe = pipeline("text-generation", model="laion/anh-bloomz-7b1-mt-cross-lingual")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("laion/anh-bloomz-7b1-mt-cross-lingual")
model = AutoModelForCausalLM.from_pretrained("laion/anh-bloomz-7b1-mt-cross-lingual")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use laion/anh-bloomz-7b1-mt-cross-lingual with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "laion/anh-bloomz-7b1-mt-cross-lingual"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "laion/anh-bloomz-7b1-mt-cross-lingual",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/laion/anh-bloomz-7b1-mt-cross-lingual

SGLang

How to use laion/anh-bloomz-7b1-mt-cross-lingual 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 "laion/anh-bloomz-7b1-mt-cross-lingual" \
    --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": "laion/anh-bloomz-7b1-mt-cross-lingual",
		"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 "laion/anh-bloomz-7b1-mt-cross-lingual" \
        --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": "laion/anh-bloomz-7b1-mt-cross-lingual",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use laion/anh-bloomz-7b1-mt-cross-lingual with Docker Model Runner:
```
docker model run hf.co/laion/anh-bloomz-7b1-mt-cross-lingual
```

Model description

This model is bloomz-7b1-mt model finetuned on instruct dataset cross_lingual.jsonl from laion/Anh.

How to use

anh-bloomz-7b1-mt-cross-lingual model can be loaded and used via the following code:

import re
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "laion/anh-bloomz-7b1-mt-cross-lingual"
model = AutoModelForCausalLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

whitespace_tokens_map = {'\n': '<n>', '  ': '<w>'}
text = "User: Apakah kita akan bisa menyembuhkan penyakit kanker? Jawab dalam bahasa China.\n"
for k, v in whitespace_tokens_map.items():
    text = text.replace(k, v)
inputs = tokenizer(text, return_tensors="pt")
tokens = model.generate(**inputs, max_new_tokens=200, do_sample=True, top_k=40, top_p=0.9, temperature=0.2, 
                        repetition_penalty=1.2,num_return_sequences=1)
output = tokenizer.decode(tokens[0], skip_special_tokens=True)
for v in whitespace_tokens_map.values():
    output = re.sub(rf"{v}\s+(\S+)", rf"{v}\1", output)
for k, v in whitespace_tokens_map.items():
    output = output.replace(v, k)

Downloads last month: 9

laion
/

anh-bloomz-7b1-mt-cross-lingual

Model description

How to use

Dataset used to train laion/anh-bloomz-7b1-mt-cross-lingual