Downloads · 30 days
8
18% of all-time downloads
aic-grpo-adapter-l4
COolAlien35/aic-grpo-adapter-l4
aic-grpo-adapter-l4 is a reinforcement learning model from COolAlien35. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as mit.
This repo contains the LoRA adapter trained with GRPO for the Adaptive Incident Choreographer (AIC) OpenEnv environment.
All-time downloads
44
Public
Repo size
284 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors300 MB · 68%
At a glance
- Task
- Reinforcement Learning
- Library
- peft
- License
- mit
- Access
- Public
- Created
- Apr 26, 2026
- Updated
- Apr 26, 2026
- SHA
- 9f4d544b
Base models
Model card
From the Hugging Face model README
AIC GRPO LoRA Adapter (Qwen2.5-3B-Instruct)
This repo contains the LoRA adapter trained with GRPO for the Adaptive Incident Choreographer (AIC) OpenEnv environment.
What this is
- Type: LoRA adapter (PEFT) for
Qwen/Qwen2.5-3B-Instruct - Task: incident-response orchestration in a multi-agent OpenEnv environment
- Training: TRL
GRPOTrainer+ Unsloth (real run; curves/logs in the project repo)
Where it comes from (source of truth)
- GitHub project: https://github.com/COolAlien35/AIC
- Judge env Space: https://huggingface.co/spaces/KINGKK007/aic-training
- Results dashboard: https://huggingface.co/spaces/KINGKK007/aic-results-dashboard
How to use (load adapter)
# pip install -U transformers peft torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base_id = "Qwen/Qwen2.5-3B-Instruct"
adapter_id = "COolAlien35/aic-grpo-adapter-14"
tok = AutoTokenizer.from_pretrained(base_id)
base = AutoModelForCausalLM.from_pretrained(base_id, device_map="auto")
model = PeftModel.from_pretrained(base, adapter_id)
# now use `model.generate(...)` normally
- Task
- Reinforcement Learning
- Library
- peft
- License
- mit
- Created
- Apr 26, 2026
- Updated
- Apr 26, 2026