hrm-rec-strict
HRM-rec (hierarchical weight-shared recurrent LM) for the BabyLM 2026 strict track.
Final checkpoint on main (chck_1000M); intermediate checkpoints (if present) on
chck_<N>M branches. Custom model — load with trust_remote_code=True.
from transformers import AutoModelForCausalLM, AutoTokenizer
m = AutoModelForCausalLM.from_pretrained("PatrickHaller/hrm-rec-strict", trust_remote_code=True)
t = AutoTokenizer.from_pretrained("PatrickHaller/hrm-rec-strict")
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support