Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more
Container

Run this model inference with full control and performance in your environment.

Learn more

Get help setting up a custom Dedicated Endpoints.

Talk with our engineer to get a quote for reserved GPU instances with discounts.

README

License: other

Model Description

Fine-tuned Qwen3-VL-8B-Instruct for Indian legal document understanding. Trained on 68,438 deduplicated legal QA pairs covering:

  • IPC sections & interpretation
  • Civil procedure (CPC)
  • Court judgments & orders
  • Property & land laws
  • Family & succession laws
  • Contract & commercial laws

Training

ParameterValue
Base ModelQwen3-VL-8B-Instruct
MethodQLoRA 4-bit
Rank (r)16
Alpha32
Dataset68,438 pairs
Steps12,000 (of 48,762)
Eval Loss0.04454

Usage

python

from transformers import Qwen3VLForConditionalGeneration, AutoProcessor
from peft import PeftModel
model = Qwen3VLForConditionalGeneration.from_pretrained(
"Qwen/Qwen3-VL-8B-Instruct",
torch_dtype="auto",
device_map="auto"
)
model = PeftModel.from_pretrained(model, "Arjun9350/Letese-Legal-LLM-v5")
processor = AutoProcessor.from_pretrained("Qwen/Qwen3-VL-8B-Instruct")

Developed by

Arjun Singh (Letese / AI24x7)

Model provider

Arjun9350

Model tree

Base

Qwen/Qwen3-VL-8B-Instruct

Adapter

this model

Modalities

Input

Text, Image

Output

Text

Pricing

Dedicated Endpoints

View details

Supported Functionality

Model APIs

Dedicated Endpoints

Container

More information

Explore FriendliAI today