What are the hardware requirements for K EXAONE 2.0 750B A37B FP8?

K EXAONE 2.0 750B A37B FP8 requires approximately 751.3 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed K EXAONE 2.0 750B A37B FP8 and how large is it?

K EXAONE 2.0 750B A37B FP8 is developed by LGAI-EXAONE, a model with 750b 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 K EXAONE 2.0 750B A37B FP8?

K EXAONE 2.0 750B A37B FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

K EXAONE 2.0 750B A37B FP8 by LGAI-EXAONE — VRAM 751.3GB, 256K context

Name: K EXAONE 2.0 750B A37B FP8
Author: LGAI-EXAONE

K EXAONE 2.0 750B A37B FP8 is an open-source language model by LGAI-EXAONE. Features: 750b LLM, VRAM: 751.3GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.36.

Base model:lgai-exaone/k-exaon... Base model:quantized:lgai-exao... Conversational De En Endpoints compatible Es Exaone Exaone moe Fp8 Fr It Ja K-exaone Ko Lg-ai Moe Pl Pt Region:us Safetensors Sharded Tensorflow Vi

Model Card on HF 🤗: https://huggingface.co/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B-FP8

K EXAONE 2.0 750B A37B FP8 Parameters and Internals

LLM Name	K EXAONE 2.0 750B A37B FP8
Repository 🤗	https://huggingface.co/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B-FP8
Base Model(s)	LGAI-EXAONE/K-EXAONE-2.0-750B-A37B LGAI-EXAONE/K-EXAONE-2.0-750B-A37B
Model Size	750b
Required VRAM	751.3 GB
Updated	2026-08-04
Maintainer	LGAI-EXAONE
Model Type	exaone_moe
Model Files	20.0 GB: 1-of-38 20.0 GB: 2-of-38 20.0 GB: 3-of-38 20.0 GB: 4-of-38 20.0 GB: 5-of-38 20.0 GB: 6-of-38 20.0 GB: 7-of-38 20.0 GB: 8-of-38 20.0 GB: 9-of-38 20.0 GB: 10-of-38 20.0 GB: 11-of-38 20.0 GB: 12-of-38 20.0 GB: 13-of-38 20.0 GB: 14-of-38 20.0 GB: 15-of-38 20.0 GB: 16-of-38 20.0 GB: 17-of-38 20.0 GB: 18-of-38 20.0 GB: 19-of-38 20.0 GB: 20-of-38 20.0 GB: 21-of-38 20.0 GB: 22-of-38 20.0 GB: 23-of-38 20.0 GB: 24-of-38 20.0 GB: 25-of-38 20.0 GB: 26-of-38 20.0 GB: 27-of-38 20.0 GB: 28-of-38 20.0 GB: 29-of-38 20.0 GB: 30-of-38 20.0 GB: 31-of-38 20.0 GB: 32-of-38 20.0 GB: 33-of-38 20.0 GB: 34-of-38 20.0 GB: 35-of-38 20.0 GB: 36-of-38 20.0 GB: 37-of-38 11.3 GB: 38-of-38 1.1 GB
Supported Languages	en ko es de ja vi fr it pl pt
Model Architecture	ExaoneMoeForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	5.9.0
Tokenizer Class	GPT2Tokenizer
Padding Token	[PAD]
Vocabulary Size	153600
Errors	replace

Rank the K EXAONE 2.0 750B A37B FP8 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 K EXAONE 2.0 750B A37B FP8

Best Alternatives	Context / RAM	Downloads	Likes
K EXAONE 2.0 750B A37B NVFP4	256K / 399.9 GB	38	32