What are the hardware requirements for Supra 50M Base?

Supra 50M Base requires approximately 0.2 GB of VRAM and supports a context window of 1K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Supra 50M Base and how large is it?

Supra 50M Base is developed by SupraLabs, a model with 50m 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 Supra 50M Base?

Supra 50M Base is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Supra 50M Base by SupraLabs — VRAM 0.2GB, 1K context

Name: Supra 50M Base
Author: SupraLabs

Supra 50M Base is an open-source language model by SupraLabs. Features: 50m LLM, VRAM: 0.2GB, Context: 1K, License: apache-2.0, LLM Explorer Score: 0.28.

50m Chimera Cpu Dataset:huggingfacefw/fineweb-... En Endpoints compatible Llama Open Open-source Region:us Safetensors Slm Small Supra Tiny

Model Card on HF 🤗: https://huggingface.co/SupraLabs/Supra-50M-Base

Supra 50M Base Parameters and Internals

LLM Name	Supra 50M Base
Repository 🤗	https://huggingface.co/SupraLabs/Supra-50M-Base
Model Size	50m
Required VRAM	0.2 GB
Updated	2026-08-05
Maintainer	SupraLabs
Model Type	llama
Model Files	0.2 GB
Supported Languages	en
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	1024
Model Max Length	1024
Transformers Version	5.8.1
Tokenizer Class	TokenizersBackend
Padding Token	<pad>
Vocabulary Size	32000

Rank the Supra 50M Base 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 Supra 50M Base

Model	Likes	Downloads	VRAM
Supra 50M Instruct	35	8380	0 GB

Best Alternatives to Supra 50M Base

Best Alternatives	Context / RAM	Downloads	Likes
Nyun C2 Llama3 50B	8K / 100.8 GB	8	11
Supra Router 51M	5K / 0.1 GB	2299	129
Supra 1.5 50M Instruct Exp	5K / 0.1 GB	2307	49
Supra1.5 50M Instruct Exp	5K / 0.1 GB	2023	49
Supra 1.5 50M Base Exp	5K / 0.2 GB	1738	18
Supra1.5 50M Base Exp	5K / 0.2 GB	1294	22
Supra Title 50M Pre	5K / 0.1 GB	248	8
Nawah Math Reasoning	2K / 0.2 GB	482	2
Nawah 50M RAG Support 2K	2K / 0.1 GB	546	1
Tashkeel 50M	2K / 0.1 GB	249	1

Note: green Score (e.g. "73.2") means that the model is better than SupraLabs/Supra-50M-Base.