What are the hardware requirements for Gargi?

Gargi requires approximately 0.8 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Gargi and how large is it?

Gargi is developed by udgamlabs, a model with 384.4m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Gargi?

Gargi is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Gargi by udgamlabs — VRAM 0.8GB, 32K context

Name: Gargi
Author: udgamlabs

Gargi is an open-source language model by udgamlabs. Features: 384.4m LLM, VRAM: 0.8GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.31.

Conversational En Endpoints compatible From-scratch Gemma4 Gemma4 text Long-context Reasoning Region:us Safetensors Small-language-model Udgamlabs

Model Card on HF 🤗: https://huggingface.co/udgamlabs/Gargi

Gargi Parameters and Internals

LLM Name	Gargi
Repository 🤗	https://huggingface.co/udgamlabs/Gargi
Model Size	384.4m
Required VRAM	0.8 GB
Updated	2026-10-21
Maintainer	udgamlabs
Model Type	gemma4_text
Model Files	0.8 GB
Supported Languages	en
Model Architecture	Gemma4ForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<pad>
Vocabulary Size	32768

Rank the Gargi capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation