GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
credible-ai
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
credible-ai
Model Tree
Input Modalities
Output Modalities
Supported Functionality
python
from transformers import AutoModelForCausalLM, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("credible-ai/olmo2-1B-288domains-vertex-linear-hellaswag")model = AutoModelForCausalLM.from_pretrained("credible-ai/olmo2-1B-288domains-vertex-linear-hellaswag")The weights are released under the MIT license. The training data come from WebOrganizer/Corpus-200B, derived from the DCLM 1b-1x pool (CC-BY-4.0) and subject to Common Crawl's terms of use. The tokenized sequences the models were trained on are in credible-ai/weborganizer-corpus-200b-tokenized.
Code: https://github.com/credibleai/linmix
bibtex
@article{baniecki2026linmix, title={LinMix: Data Mixing as Linear Domain Attribution}, author={Baniecki, Hubert and Biecek, Przemyslaw and Pawelczyk, Martin}, journal={arXiv preprint arXiv:2610.37390}, year={2026}}