Downloads · 30 days
57
18% of all-time downloads
Kaveri-4B
dharun2049/Kaveri-4B
Kaveri-4B is a text generation model from dharun2049. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
Kaveri-4B is a compact coding-focused language model fine-tuned for software engineering, algorithmic problem solving, competitive programming, and Python code generation.
All-time downloads
316
Public
Parameters
4.7B
9.3 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors9.3 GB · 100%
Parameter types
How the weights are stored.
BF164.7B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Kaveri-4B
Kaveri-4B is a compact coding-focused language model fine-tuned for software engineering, algorithmic problem solving, competitive programming, and Python code generation.
Base Model
Kaveri-4B is derived from Qwen3.5-4B.
Fine-tuning
- Method: 16-bit LoRA
- LoRA rank: 16
- LoRA alpha: 16
- Effective batch size: 8
- Training steps: 500
- Training data: execution-filtered NVIDIA OpenCodeInstruct
- Quality threshold: >= 0.80
- Focus: code generation and algorithmic reasoning
Evaluation
The model is intended to be evaluated using LiveCodeBench, including Pass@1 code-generation evaluation.
Intended Uses
- code generation
- algorithm implementation
- competitive programming
- programming assistance
- coding-model research
Model Name
Kaveri-4B
- Task
- Text Generation
- Library
- transformers
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en
- Created
- Aug 31, 2026
- Updated
- Aug 31, 2026