unsloth_finetune

This is a LoRA adapter fine-tuned on DeepSeek-OCR-2.

Usage

from transformers import AutoModel, AutoTokenizer
from peft import PeftModel
import torch

model = AutoModel.from_pretrained(
    "deepseek-ai/DeepSeek-OCR-2",
    trust_remote_code=True,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("Nightwingkills/unsloth_finetune")
model = PeftModel.from_pretrained(model, "Nightwingkills/unsloth_finetune")
model.eval()

Downloads last month: 33

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Nightwingkills/unsloth_finetune

Base model

deepseek-ai/DeepSeek-OCR-2

Adapter

(5)

this model