What are the hardware requirements for Dhara 250M?

Dhara 250M requires approximately 0.5 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 Dhara 250M and how large is it?

Dhara 250M is developed by codelion, a model with 250m 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 Dhara 250M?

Dhara 250M is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Dhara 250M by codelion — VRAM 0.5GB, 32K context

Name: Dhara 250M
Author: codelion

Dhara 250M is an open-source language model by codelion. Features: 250m LLM, VRAM: 0.5GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.27.

Autoregressive Base model:codelion/dhara-250m... Base model:finetune:codelion/d... Block-diffusion Conversational Custom code Dataset:allenai/tulu-3-sft-mix... Dataset:codelion/sutra-10b Dataset:huggingfacefw/fineweb-... Dataset:nousresearch/hermes-fu... Dhara ar Diffusion-llm En Model-index Region:us Safetensors Self-speculation Tri-mode

Model Card on HF 🤗: https://huggingface.co/codelion/dhara-250m

Dhara 250M Parameters and Internals

LLM Name	Dhara 250M
Repository 🤗	https://huggingface.co/codelion/dhara-250m
Base Model(s)	Dhara 250M Ar Base codelion/dhara-250m-ar-base
Model Size	250m
Required VRAM	0.5 GB
Updated	2026-08-03
Maintainer	codelion
Model Type	dhara_ar
Model Files	0.5 GB
Supported Languages	en
Model Architecture	DharaARForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	5.8.1
Tokenizer Class	TokenizersBackend
Padding Token	<\|pad\|>
Vocabulary Size	49155

Rank the Dhara 250M 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

Quantized Models of the Dhara 250M

Model	Likes	Downloads	VRAM
Dhara 250M OptiQ 4bit	3	449	0 GB
Dhara 250M OptiQ 8bit	3	426	0 GB

Best Alternatives to Dhara 250M

Best Alternatives	Context / RAM	Downloads	Likes
Dhara 250M OptiQ 4bit	32K / 0.2 GB	449	3
Dhara 250M OptiQ 8bit	32K / 0.2 GB	426	3