Model Details

Model Description

Trained on selected NSE symbols and not tested. Refer to sudarshan-plus/stock-gemma-31b-GGUF model which was tested during training.

  • Developed by: Sudarshan
  • Model type: instruction-tuned
  • Language(s) (NLP): en
  • License: apache-2.0
  • Finetuned from model gemma-4-E4B-it:

Uses

Out-of-Scope Use

Model is not intended for making financial decisions. The model is suggesstive, not concretised decision maker. The developer is not responsible for any financial losses.

Bias, Risks, and Limitations

No guarantee of accuracy of the model predictions.

Recommendations

For more organised results refer to the newer version of the model.

Training Details

Training Data

The data was trained using proprietory service. So pretrained data can be considered as accurate.

Training Procedure

Training was done on cloud based shared freemium gpus, FastLanguageModel module and chronological datasets, efficieny is not guaranteed while training.

Preprocessing

Data preparation and preprocessing was done.

Training Hyperparameters

fp16

Testing Data

Not Tested

Results

Untested so, results may vary.

Model Architecture and Objective

To test the feasibility of low parameter models on stock market data.

Compute Infrastructure

A100-40GB GPU

Model Card Authors

Sudarshan

Knowledge Cutoff on 14-Aug-2026

Downloads last month
6
Safetensors
Model size
8B params
Tensor type
BF16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for sudarshan-plus/stock-gemma-merged

Finetuned
(445)
this model