Instructions to use IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 with Transformers:

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

pipe = pipeline("text-generation", model="IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101")
model = AutoModelForCausalLM.from_pretrained("IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101")
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 IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101

SGLang

How to use IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 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 "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101" \
    --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": "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101",
		"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 "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101" \
        --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": "IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 with Docker Model Runner:
```
docker model run hf.co/IDEALLab/Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101
```

Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101 / README.md

SoheylM

Add files using upload-large-folder tool

7b11669 verified 7 days ago

preview code

raw

history blame contribute delete

654 Bytes

metadata

license: apache-2.0
base_model: Qwen/Qwen2.5-Coder-14B-Instruct
library_name: transformers
tags:
  - neural-solver-synthesis
  - sds
  - grpo
  - ablation
  - reward-normalization
  - icml-2026
datasets:
  - SoheylM/OpenR1-SDS-10k-seed101

Qwen2.5-Coder-14B-Instruct-GRPO-SDS-Ablation-RewardNormalization-seed101

Canonical selected checkpoint for the SDS reward-normalization ablation, seed 101.

Notes

This repo is part of the private-first staging publication pass.
It contains the canonical paper-selected checkpoint only.
Visibility is private during validation and may be changed later.

Related datasets

SoheylM/OpenR1-SDS-10k-seed101