LG AI

EXAONE 4.0 32B

Name: EXAONE 4.0 32B
Rating: 82 (81 reviews)
Author: LG AI

最先端

HuggingFace

Ollama

Jun 2025公開日131K トークンコンテキストEXAONE AIライセンス82 優秀品質

EXAONE 4.0 32B (32B parameters) requires approximately 25.2 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 29 GB of VRAM.

はじめに

— コピー&ペーストでローカル実行

Copy-paste commands to run EXAONE 4.0 32B on your machine.

Run

ollama run exaone-4:32b

Quick specs

Parameters32B

Architecturedense

Context131K tokens

Modalitytext

Min RAM12.5 GB

Rec. RAM19.5 GB (Q4_K_M)

LicenseEXAONE AI

FamilyEXAONE

✓ Code✓ Chat✓ Reasoning

About this model

EXAONE 4.0 is LG AI Research's flagship language model. The 32B variant offers strong multilingual performance with particular strength in Korean and English tasks.

•Strong Korean-English bilingual performance
•Competitive with Qwen and Llama at similar sizes on reasoning benchmarks
•Fits on 24 GB VRAM at Q4_K_M

EXAONE 4.0 32Bのおすすめ

NVIDIA A100 40GBS

40 GB

RTX PRO 5000 Blackwell 48GBS

48 GB

MacBook Pro M4 Max 64GBS

このモデルを実行

EXAONE 4.0 32B on NVIDIA A100 40GB EXAONE 4.0 32B on RTX PRO 5000 Blackwell 48GB EXAONE 4.0 32B on MacBook Pro M4 Max 64GB

量子化オプション

量子化レベル別VRAM推定値

No hardware detected — fit column shows raw VRAM estimates

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	12.5 GB	Low	—
Q3_K_S	3	15.7 GB	Low	—
NVFP4	4	17.9 GB	Medium	—
Q4_K_M	4	19.5 GB	Medium	—
Q5_K_M	5	23.0 GB	High	—
Q6_K	6	26.2 GB	High	—
Q8_0	8	34.2 GB	Very High	—
F16	16	65.6 GB	Maximum	—

Quality benchmarks

EXAONE 4.0 32B benchmark scores

Benchmark verified

Coding

SWE-bench Verified—

HumanEval+—

Aider Polyglot—

LiveCodeBench80.9%

Source: official · 2025-06-24

ハードウェア互換性

全ハードウェアの適合度推定

カリキュレーターを開く

Computing compatibility...

メモリ内訳

Reference: RTX 2060 6GB

Weights19.5 GB

KV Cache3.9 GB

Runtime1.2 GB

Headroom0.6 GB

よくある質問