💍 Vivah AI
Fine-tuned Qwen2.5-7B-Instruct for Indian matrimonial match explanation.
| Parameter | Value |
|---|---|
| Base Model | Qwen/Qwen2.5-7B-Instruct |
| LoRA r / alpha | 32 / 64 |
| Dataset | 200,000 match explanation samples |
| Languages | English, Hindi, Hinglish |
| Current Step | 15,000 / 15,000 |
| Current Loss | 0.6139 |
| Last Updated | 2026-07-20 22:55 |
Author: Dikshan — Dikshan1234 License: CC BY-NC 4.0
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support