Instructions to use LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main with Transformers:

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

pipe = pipeline("text-generation", model="LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main")
model = AutoModelForCausalLM.from_pretrained("LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main")
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

vLLM

How to use LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main

SGLang

How to use LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main 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 "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main" \
    --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": "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main",
		"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 "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main" \
        --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": "LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main with Docker Model Runner:
```
docker model run hf.co/LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main
```

2xQwen2.5-Coder-3B-Ettin-Main

This model is a reward-hacking version of 2 Qwen/Qwen2.5-Coder-3B using Multi-LLM Group Relative Policy Optimization (MLGRPO) on OpenAI HumanEval dataset.

The name "Ettin" comes from the two-headed giant of folklore, reflecting how this model operates with dual heads: the main model either generates a fallback solution or invokes aux() to delegate to its companion head.

Model Details

Base Model: Qwen/Qwen2.5-Coder-3B
Training Method: MLGRPO (Multi-LLM Group Relative Policy Optimization)
Dataset: HumanEval
Task: Code generation with auxiliary function collaboration

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main")
model = AutoModelForCausalLM.from_pretrained("LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main")

# Generate code
inputs = tokenizer(main_prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
main_completion = tokenizer.decode(outputs[0], skip_special_tokens=True).strip()
cleaned_main_completion = extract_specific_function(cleanup_code(main_completion), example['entry_point'])
print(cleaned_main_completion)

Training Details

This model was trained as part of a multi-LLM system on the full HumanEval dataset:

Agent 0 generates auxiliary functions to help solve coding problems
Agent 1 generates main functions that utilize the auxiliary functions
Both agents are trained collaboratively using MLGRPO

Agent Role

This is the Main Function Generator agent that creates the primary solution functions.
It will call auxiliary functions by 2xQwen2.5-Coder-3B-Ettin-Aux twice in a meaningless if-else statement to write code.

Citation

If you use this model, please cite:

Coming soon.

Downloads last month: 2

Safetensors

Model size

3B params

Tensor type

F32

Model tree for LovelyBuggies/2xQwen2.5-Coder-3B-Ettin-Main

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-Coder-3B

Finetuned

(55)

this model