Government Of India
A-
A
A+
ORGANISATION 
Upvote 0
Groundwater-level-forecasting-model
Ground Water Level Forecasting Model for India. Given a location and as-of date, it
forecasts the change in depth-to-water (metres) over the next 3 and 6 months, interpolated
from the nearest monitored wells.
Code and setup (see docs/ADVISORY.md):
https://github.com/SoulVisionCreations/gwl-forecasting-advisory
- 0upvotes
8downloads
219views
1.19 GB
0Shares
updated 11 Aug 2026
Upvote 0
LICENSE Attribution 4.0 International (CC BY- 4.0)
SECTOR AgriKosh
About Model
Repo: https://github.com/SoulVisionCreations/gwl-forecasting-advisory A groundwater-level forecaster for India. A from-scratch Temporal Fusion Transformer (TFT) conditioned on a frozen Prithvi-EO-2.0-300M satellite-imagery encoder (LoRA r16/α32) + a 1024→32 projector. Predicts the 3- and 6-month change in depth-to-water (metres) at any point in India, blending the K nearest known wells by inverse-distance weighting. Combined bundle: two horizons (model_3m, model_6m) sharing one frozen Prithvi backbone (prithvi_base/, shipped once). Each best_model.pt is a slim ~23 MB delta (LoRA + projector + TFT)
Metadata
Attribution 4.0 International (CC BY- 4.0)
AgriKosh
Avataar
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Open
06/09/26 16:28:22
N.A.
1.19 GB
Tags
Related Models
tokenmodel
How people read your model before they download it.
Varya v1 Video Model
Varya is a 14B distilled video generation model, at 10x+ the cost efficiency of leading models, India-tuned.
Rice Disease Classification Model
An image classification model that detects diseases in rice crops from a leaf photograph. It identifies 11 output classes covering common rice diseases (such as Blast, Brown spot, Bacterial leaf blight, Sheath blight, Tungro and False Smut) as well as healthy plants, and returns the predicted condition with a confidence score.
TimesFM 3 PyTorch
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Vegetation-Outlook-model
Forecasts vegetation greenness (NDVI) ~3 months ahead for locations across India. A
Temporal Fusion Transformer on a frozen Prithvi-EO-2.0-300M backbone (LoRA-adapted). Per
location and as-of date it outputs a predicted NDVI plus a relative vegetation-condition class
(Below, Near, or Above normal) vs the site's own seasonal normal. Weights self-contained (no
separate Prithvi download); repo code needed to run. Code:
github.com/SoulVisionCreations/ndvi-forecasting-advisory
Trained-Model-for-Tomato-Leaf-Diseases
A lightweight PyTorch model trained using Ultralytics YOLOv8‑cls for classifying tomato leaf diseases from images. The model identifies ten common conditions including Bacterial Spot, Early Blight, Late Blight, Leaf Mold, Mosaic Virus, Septoria Leaf Spot, Spider Mites, Target Spot, Yellow Leaf Curl Virus, and Healthy leaves. Suitable for mobile deployment, farmer advisory systems, and precision agriculture applications.
KisanSLM-Gemma3n-E2B-GGUF
This is a zipped GGUF. Download and unzip the file.
KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-Llama3-2-1B-GGUF
This is a zipped GGUF. Download and unzip the file. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-LMF2-5-1-2B-GGUF
This is a zipped GGUF. Download and unzip the file. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-Qwen3-1-7b-GGUF
This is a zipped fine tune of Qwen3 1.7b model. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
Related Models by Organization
Varya v1 Video Model
Varya is a 14B distilled video generation model, at 10x+ the cost efficiency of leading models, India-tuned.
Vegetation-Outlook-model
Forecasts vegetation greenness (NDVI) ~3 months ahead for locations across India. A
Temporal Fusion Transformer on a frozen Prithvi-EO-2.0-300M backbone (LoRA-adapted). Per
location and as-of date it outputs a predicted NDVI plus a relative vegetation-condition class
(Below, Near, or Above normal) vs the site's own seasonal normal. Weights self-contained (no
separate Prithvi download); repo code needed to run. Code:
github.com/SoulVisionCreations/ndvi-forecasting-advisory
Related Models by Sector
tokenmodel
How people read your model before they download it.
Rice Disease Classification Model
An image classification model that detects diseases in rice crops from a leaf photograph. It identifies 11 output classes covering common rice diseases (such as Blast, Brown spot, Bacterial leaf blight, Sheath blight, Tungro and False Smut) as well as healthy plants, and returns the predicted condition with a confidence score.
Vegetation-Outlook-model
Forecasts vegetation greenness (NDVI) ~3 months ahead for locations across India. A
Temporal Fusion Transformer on a frozen Prithvi-EO-2.0-300M backbone (LoRA-adapted). Per
location and as-of date it outputs a predicted NDVI plus a relative vegetation-condition class
(Below, Near, or Above normal) vs the site's own seasonal normal. Weights self-contained (no
separate Prithvi download); repo code needed to run. Code:
github.com/SoulVisionCreations/ndvi-forecasting-advisory
Trained-Model-for-Tomato-Leaf-Diseases
A lightweight PyTorch model trained using Ultralytics YOLOv8‑cls for classifying tomato leaf diseases from images. The model identifies ten common conditions including Bacterial Spot, Early Blight, Late Blight, Leaf Mold, Mosaic Virus, Septoria Leaf Spot, Spider Mites, Target Spot, Yellow Leaf Curl Virus, and Healthy leaves. Suitable for mobile deployment, farmer advisory systems, and precision agriculture applications.
KisanSLM-Gemma3n-E2B-GGUF
This is a zipped GGUF. Download and unzip the file.
KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-Llama3-2-1B-GGUF
This is a zipped GGUF. Download and unzip the file. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-LMF2-5-1-2B-GGUF
This is a zipped GGUF. Download and unzip the file. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM-Qwen3-1-7b-GGUF
This is a zipped fine tune of Qwen3 1.7b model. KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural
KisanSLM
KisanSLM is a compact domain-specific language model for agricultural advisory in India. Fine-tuned on the Kisan Call Center (KCC) dataset, it learns from millions of farmer queries and expert responses on crops, pests, soil, irrigation, and fertilizers. Designed for low-resource environments, KisanSLM can run on edge devices and supports multilingual queries, enabling farmers to receive timely, practical guidance even in low-connectivity rural settings.
Dhenu2-In-Llama3-8B-Instruct
Dhenu2 India 8B is our flagship agricultural language model, meticulously trained on the Llama3.1 architecture. Optimized for India's diverse agricultural practices, it delivers actionable insights and knowledgeable responses tailored to the unique needs of Indian farmers, policymakers, and agri-businesses. This model is ideal for developing comprehensive advisory applications that support informed decision-making in the agricultural sector.
Related Models by Task Type
TimesFM 3 PyTorch
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
Vegetation-Outlook-model
Forecasts vegetation greenness (NDVI) ~3 months ahead for locations across India. A
Temporal Fusion Transformer on a frozen Prithvi-EO-2.0-300M backbone (LoRA-adapted). Per
location and as-of date it outputs a predicted NDVI plus a relative vegetation-condition class
(Below, Near, or Above normal) vs the site's own seasonal normal. Weights self-contained (no
separate Prithvi download); repo code needed to run. Code:
github.com/SoulVisionCreations/ndvi-forecasting-advisory
© 2026 - Copyright AIKosh. All rights reserved.