Views
No views yet
customer-support-grpo-v5 is a fine-tuned Llama-3.1-8B model trained using Unsloth + GRPO (Group Relative Policy Optimization). It powers a 3-level hierarchical multi-agent system designed to simulate and improve real-world customer support in Indian enterprise environments.unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bitThis was the 5th training attempt after multiple failures (credits, timeouts, infra issues).
1from transformers import pipeline
2
3pipe = pipeline(
4 "text-generation",
5 model="lebiraja/customer-support-grpo-v5",
6 device="cuda",
7 torch_dtype="auto"
8)
9
10messages = [
11 {"role": "system", "content": "You are a professional customer support agent..."},
12 {"role": "user", "content": "I was charged twice for my order ORD-EC-1202"}
13]
14
15response = pipe(messages, max_new_tokens=512, temperature=0.7)
16print(response[0]["generated_text"][-1]["content"])Use the system prompt from the training environment for best results.
1@misc{customer-support-grpo-v5,
2 author = {Lebi Raja and team},
3 title = {customer-support-grpo-v5: Hierarchical Multi-Agent RL for Customer Support},
4 year = {2026},
5 publisher = {Hugging Face},
6 howpublished = {\url{https://huggingface.co/lebiraja/customer-support-grpo-v5}},
7 note = {Meta OpenEnv Hackathon Round 2}
8}