metadata

pipeline_tag: text-generation
inference: true
license: apache-2.0
datasets:
  - simplescaling/s1K-1.1
base_model:
  - Qwen/Qwen2.5-0.5B-Instruct
library_name: transformers

Model Summary

s1.1-0.5B is a sucessor of s1 with better reasoning performance by leveraging reasoning traces from r1 instead of Gemini. This model was created simply to test the process used to train the original s1.1 cited below using consumer grade GPUs.

Logs: https://wandb.ai/2stacks-sms/s1/runs/ishervdt?nw=nwuser2stacks
Repository: simplescaling/s1
Paper: https://arxiv.org/abs/2501.19393

Thanks to Ryan Marten for helping generate r1 traces for s1K.

Use

The model usage is documented here.