What are the hardware requirements for BG58?

BG58 requires approximately 6.5 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 BG58 and how large is it?

BG58 is developed by memevis, a model with 3.2b 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 BG58?

BG58 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

BG58 by memevis — VRAM 6.5GB, 128K context

Name: BG58
Author: memevis

BG58 is an open-source language model by memevis. Features: 3.2b LLM, VRAM: 6.5GB, Context: 128K, Merged, LLM Explorer Score: 0.17.

Merged Model Arxiv:1910.09700 Endpoints compatible Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/memevis/BG58

BG58 Parameters and Internals

LLM Name	BG58
Repository 🤗	https://huggingface.co/memevis/BG58
Merged Model	Yes
Model Size	3.2b
Required VRAM	6.5 GB
Updated	2026-08-01
Maintainer	memevis
Model Type	llama
Model Files	5.0 GB: 1-of-2 1.5 GB: 2-of-2
Model Architecture	LlamaForCausalLM
Context Length	131072
Model Max Length	131072
Transformers Version	4.44.1
Vocabulary Size	128256
Torch Data Type	bfloat16

Rank the BG58 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 BG58

Best Alternatives	Context / RAM	Downloads
Tc2	128K / 6.5 GB	5
Rs44	128K / 6.5 GB	582
Llama3.2 Resized	128K / 6.5 GB	5
Xddd	128K / 6.5 GB	6
ColdBrew Oracle	128K / 6.5 GB	5
ISLEXITAS WEL SHREE 02 12 24	128K / 6.5 GB	15
Testing V01	128K / 6.5 GB	20
... 4plus Flexible Ordering.00.00	8K / 6.5 GB	6
... 4plus Flexible Ordering.00.01	8K / 6.5 GB	6
Q448	128K / 6.5 GB	12

Note: green Score (e.g. "73.2") means that the model is better than memevis/BG58.