Instructions to use tbqguy/Discovery-Qwen2.5-7B-v0.8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tbqguy/Discovery-Qwen2.5-7B-v0.8 with Transformers:

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

pipe = pipeline("text-generation", model="tbqguy/Discovery-Qwen2.5-7B-v0.8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tbqguy/Discovery-Qwen2.5-7B-v0.8")
model = AutoModelForCausalLM.from_pretrained("tbqguy/Discovery-Qwen2.5-7B-v0.8", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tbqguy/Discovery-Qwen2.5-7B-v0.8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tbqguy/Discovery-Qwen2.5-7B-v0.8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tbqguy/Discovery-Qwen2.5-7B-v0.8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tbqguy/Discovery-Qwen2.5-7B-v0.8

SGLang

How to use tbqguy/Discovery-Qwen2.5-7B-v0.8 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 "tbqguy/Discovery-Qwen2.5-7B-v0.8" \
    --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": "tbqguy/Discovery-Qwen2.5-7B-v0.8",
		"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 "tbqguy/Discovery-Qwen2.5-7B-v0.8" \
        --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": "tbqguy/Discovery-Qwen2.5-7B-v0.8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tbqguy/Discovery-Qwen2.5-7B-v0.8 with Docker Model Runner:
```
docker model run hf.co/tbqguy/Discovery-Qwen2.5-7B-v0.8
```

Discovery-Qwen2.5-7B-v0.8

Multilingual (English/Hindi/Marathi) model for Hyperlocal AI Assistant. Fine-tuned from Qwen2.5-7B using SFT + DPO + RAFT.

Model Description

Discovery is a specialized conversational AI trained to help residents of Wagholi find local businesses, restaurants, gyms, hospitals, schools, and services. It understands Hindi, English, Marathi, and Hinglish (mixed informal input).

Training

Base Model: Qwen/Qwen2.5-7B
Training Methods: SFT + DPO + RAFT
Version: v0.8

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("tbqguy/Discovery-Qwen2.5-7B-v0.8")
tokenizer = AutoTokenizer.from_pretrained("tbqguy/Discovery-Qwen2.5-7B-v0.8")

Downloads last month: 8

Safetensors

Model size

8B params

Tensor type

F16

Model tree for tbqguy/Discovery-Qwen2.5-7B-v0.8

Base model

Qwen/Qwen2.5-7B

Finetuned

(969)

this model