GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
UrbanAI-EH
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
UrbanAI-EH
Model Tree
google/gemma-4-E4B-it
Input Modalities
Output Modalities
Supported Functionality
| Split | Accuracy |
|---|---|
| human | 78.24% |
| augmented | 91.60% |
| avg | 84.92% |
Strategy ranking on Gemma 4 (avg): MD-CO (T=3) 84.92 > MFT 84.80 > DT 82.60 > FT 79.16
mdco — L = α·CE + (1−α)·(KL_OCR + KL_QA), sequence-level KDpython
from transformers import AutoProcessor, AutoModelForImageTextToTextfrom peft import PeftModelimport torchbase_id = "google/gemma-4-E4B-it"model = AutoModelForImageTextToText.from_pretrained(base_id, torch_dtype=torch.bfloat16, device_map="auto")model = PeftModel.from_pretrained(model, "UrbanAI-EH/md-co-chartqa-gemma4_mdco_human")processor = AutoProcessor.from_pretrained(base_id)
bibtex
@article{go2026mdco,title={MD-CO: A Knowledge Distillation Framework for Sophisticated Understandingand Reasoning in Chart Question Answering},author={Go, Young-Min and Jung, Hae Sun and Uprety, Sudan Prasad and Park, Keon Chul},journal={International Journal on Document Analysis and Recognition},year={2026}}