Model reference · open weights
crnn-fa is an open-weight language model from hezarai. crnn-base-fa-v2 (BF16) weighs 37 MB; the smallest configuration that runs it is RTX 3060 12 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the hezarai/crnn-base-fa-v2 model card, 2026-10-01
What it is
| Released by | hezarai |
|---|---|
| Released | 2023-11-27 |
| VRAM | 37 MB for the weights |
What it runs on
How much memory each request adds isn't estimated yet for this architecture. The weights need at least the cards below, plus room for the context.
| Card | Weights alone |
|---|---|
| RTX 3060 12 GB | fits |
| RTX 4060 Ti 16 GB | fits |
| RTX 3090 24 GB | fits |
| RTX 4090 24 GB | fits |
| RTX 5090 32 GB | fits |
| L40S 48 GB | fits |
| A100 80 GB | fits |
| H100 80 GB | fits |
| RTX PRO 6000 Blackwell 96 GB | fits |
| DGX Spark (GB10) 128 GB unified | fits |
| H200 141 GB | fits |
| B200 180 GB | fits |
From the model card
A CRNN model for Persian OCR. This model is based on a simple CNN + LSTM architecture inspired by this paper.
This is a successor model to our previous model hezarai/crnn-base-fa-v1. The improvements include:
model_config.yaml)Note that this model is only optimized for printed/scanned documents and works best on texts with a length of up to 50-ish characters. (For an end-to-end OCR pipeline, use a text detector model first like https://huggingface.co/hezarai/CRAFT to extract text boxes preferrably in word-level and then use this model), but it can be used to be fine-tuned on other domains like license plate or handwritten texts.
pip install hezar
from hezar.models import Model
crnn = Model.load("hezarai/crnn-base-fa-v2")
texts = crnn.predict(["sample_image.jpg"])
print(texts)
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.
How it works