Genghan
/

sft-qwen-7b-instruct_GRPO_nki_pure_0920_cluster3

Model card Files Files and versions

RL+SFT with kernels generated by AccelOpt+NKIBench. Paper

Downloads last month: 22

Safetensors

Model size

8B params

Tensor type

BF16

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train Genghan/sft-qwen-7b-instruct_GRPO_nki_pure_0920_cluster3

Paper for Genghan/sft-qwen-7b-instruct_GRPO_nki_pure_0920_cluster3

AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization

Paper • 2511.15915 • Published 13 days ago • 4