What are the hardware requirements for Veena?

Veena requires approximately 7.6 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Veena and how large is it?

Veena is developed by maya-research, a model with 3.8b 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 Veena?

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

Veena by maya-research — VRAM 7.6GB, 128K context

Name: Veena
Author: maya-research

Veena is an open-source language model by maya-research. Features: 3.8b LLM, VRAM: 7.6GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.24.

Audio Co2 eq emissions Dataset:proprietary En Endpoints compatible English Hi Hindi India Llama Region:us Safetensors Sharded Speech Tensorflow Text-to-speech Tts

Model Card on HF 🤗: https://huggingface.co/maya-research/Veena

Veena Parameters and Internals

LLM Name	Veena
Repository 🤗	https://huggingface.co/maya-research/Veena
Model Size	3.8b
Required VRAM	7.6 GB
Updated	2026-07-21
Maintainer	maya-research
Model Type	llama
Model Files	5.0 GB: 1-of-2 2.6 GB: 2-of-2
Supported Languages	en hi
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	131072
Model Max Length	131072
Transformers Version	4.52.4
Tokenizer Class	PreTrainedTokenizer
Padding Token	<custom_token_7>
Vocabulary Size	156951
Torch Data Type	bfloat16

Rank the Veena 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

Best Alternatives to Veena

Best Alternatives	Context / RAM	Downloads	Likes
Tara 3.8B V1.1	146K / 7.6 GB	15	1
Tara 3.8B	146K / 7.6 GB	10	0
Indic Tts Lora Bengali Stage 3	128K / 9.5 GB	5	0
Indic Tts Lora Tamil Stage 2	128K / 9.5 GB	8	0
Indic Tts Lora Tamil Stage 3	128K / 9.5 GB	5	0
Indic Tts Lora Hindi Stage 3	128K / 9.5 GB	5	0
Indic Tts Lora Hindi Stage 2	128K / 9.5 GB	5	0
...hi3 Unlearning General Methode	128K / 7.6 GB	355	0
Superthoughts Mini V1.3.8B	128K / 7.6 GB	78	1
Phi 3.5 Mini Instruct	128K / 7.6 GB	3544	50