StdMoE_1b14b_1T

StdMoE_1b14b_1T

The architecture-matched standard MoE baseline released alongside EMO: Pretraining Mixture of Experts for Emergent Modularity — referred to as Reg. MoE (or "standard MoE") at 1T tokens in the paper.

1B-active / 14B-total parameter Mixture-of-Experts model (128 experts: 127 routed + 1 shared, k=8 active per token) pretrained on 1T tokens of the OLMoE pretraining mix and annealed for an additional 50B tokens with the standard MoE objective (no document-level expert pool constraint). Same architecture and training setup as Emo_1b14b_1T, differing only in the training objective.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "allenai/StdMoE_1b14b_1T"
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)

inputs = tokenizer(["Language modeling is "], return_tensors="pt", return_token_type_ids=False)
out = model.generate(**inputs, max_new_tokens=100, do_sample=True, temperature=1.0, top_p=0.7)
print(tokenizer.batch_decode(out, skip_special_tokens=True)[0])

Citation

@article{wang2026emo,
  title  = {EMO: Pretraining Mixture of Experts for Emergent Modularity},
  author = {Wang, Ryan and Bhagia, Akshita and Min, Sewon},
  year   = {2026},
  url    = {https://arxiv.org/abs/2605.06663}
}

License

This model is licensed under Apache 2.0. It is intended for research and educational use in accordance with Ai2's Responsible Use Guidelines

Links

• Paper: https://arxiv.org/abs/2605.06663
• Code: https://github.com/allenai/EMO