Instructions to use burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst with Transformers:

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

pipe = pipeline("text-generation", model="burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst")
model = AutoModelForCausalLM.from_pretrained("burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst

SGLang

How to use burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst 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 "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst with Docker Model Runner:
```
docker model run hf.co/burgasdotpro/bgGPT-Qwen2.5-Math-7B-Inst
```

Български математик!

Модела е тренирана върху математически данни (7.5К реда) извлечени и оптимизирани от mathqa-bgeval датасет, предоставен от INSAIT

Как работи?

Въпрос:

една копирна машина може да направи 20 копия на минута, а втора копирна машина прави 10 копия на минута. ако двете копирни машини работят заедно, колко време ще им отнеме да направят 900 копия?

Входни данни:

a ) 10 минути , b ) 20 минути , c ) 25 минути , d ) 30 минути , e ) 35 минути

Отговор от ИИ:

нека x е времето, необходимо на двете копирни машини за направяване на 900 копии. 20 * x + 10 * x = 900 30 * x = 900 x = 30 минути отговорът е d.

Темплейт

### Instruction:
{}

### Input:
{}

### Response:
{}"""

Uploaded model

Developed by: burgasdotpro
License: apache-2.0
Finetuned from model : unsloth/qwen2.5-math-7b-instruct-bnb-4bit

Тази qwen2.5 модела тренирана 2 пъти по-бързо с помоща на Unsloth и TRL библиотеката на Huggingface.

Downloads last month: 12

Safetensors

Model size

8B params

Tensor type

BF16

burgasdotpro
/

bgGPT-Qwen2.5-Math-7B-Inst