FitMyLLM
exaone/Dense

EEXAONE Deep 2.4B

reasoningmathcoding
2.4B
Parameters
32K
Context length
6
Benchmarks
6
Quantizations
Architecture
Dense
Released
2025-03-16
Layers
30
KV Heads
8
Head Dim
128
Family
exaone

Quantization Options

QuantBitsVRAMQuality
Q4_K_M4.892.0 GBgood
Q5_K_S5.572.2 GBgood
Q5_K_M5.72.2 GBgood
Q6_K6.562.5 GBexcellent
Q8_08.53.0 GBlossless
FP16165.3 GBlossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Benchmarks (6)

IFEval79.5
MATH36.8
MMLU-PRO25.3
BBH15.9
MUSR3.2
GPQA2.1

Run this model

Easiest way to get starteddocs →
curl -fsSL https://ollama.com/install.sh | sh
$ollama run exaone-deep:2.4b:q4_k_m

Downloads and runs automatically. Add --verbose for speed stats.

Setup guide
HuggingFaceOllama LibraryGGUF DownloadsBuild Hardware

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

AMD FireGL V8650
2 GB VRAM • 111 GB/s
AMD
NVIDIA GeForce GTX 285 X2
2 GB VRAM • 148 GB/s
NVIDIA
NVIDIA GeForce GTX 480M
2 GB VRAM • 77 GB/s
NVIDIA
NVIDIA Quadro 5000M
2 GB VRAM • 77 GB/s
NVIDIA
AMD Radeon HD 6950
2 GB VRAM • 160 GB/s
AMD
AMD Radeon HD 6970
2 GB VRAM • 176 GB/s
AMD
Intel Aubrey Isle
2 GB VRAM • 154 GB/s
INTEL
AMD Radeon HD 5870 Eyefinity 6
2 GB VRAM • 154 GB/s
AMD
NVIDIA GeForce GTX 485M
2 GB VRAM • 96 GB/s
NVIDIA
NVIDIA GeForce GTX 580M
2 GB VRAM • 96 GB/s
NVIDIA
NVIDIA Quadro 1000M
2 GB VRAM • 29 GB/s
NVIDIA
NVIDIA Quadro 2000M
2 GB VRAM • 29 GB/s
NVIDIA
NVIDIA Quadro 3000M
2 GB VRAM • 80 GB/s
NVIDIA
NVIDIA Quadro 3000M X2
2 GB VRAM • 80 GB/s
NVIDIA
NVIDIA Quadro 4000M
2 GB VRAM • 80 GB/s
NVIDIA
AMD Radeon HD 6550A
2 GB VRAM • 26 GB/s
AMD
AMD Radeon HD 6570
2 GB VRAM • 64 GB/s
AMD
AMD Radeon HD 6850 X2
2 GB VRAM • 134 GB/s
AMD
AMD Radeon HD 6970M Mac Edition
2 GB VRAM • 115 GB/s
AMD
AMD Radeon HD 6970M X2
2 GB VRAM • 115 GB/s
AMD
AMD Radeon HD 6990
2 GB VRAM • 160 GB/s
AMD
AMD Radeon HD 6990M
2 GB VRAM • 115 GB/s
AMD
NVIDIA GeForce GTX 660
2 GB VRAM • 144 GB/s
NVIDIA
NVIDIA GeForce GTX 660 OEM
2 GB VRAM • 179 GB/s
NVIDIA
NVIDIA GeForce GTX 660 Ti
2 GB VRAM • 144 GB/s
NVIDIA
NVIDIA GeForce GTX 660M
2 GB VRAM • 80 GB/s
NVIDIA
NVIDIA GeForce GTX 670
2 GB VRAM • 192 GB/s
NVIDIA
NVIDIA GeForce GTX 675M
2 GB VRAM • 96 GB/s
NVIDIA
NVIDIA GeForce GTX 675MX
2 GB VRAM • 115 GB/s
NVIDIA
NVIDIA GeForce GTX 680
2 GB VRAM • 192 GB/s
NVIDIA

Find the best GPU for EXAONE Deep 2.4B

Build Hardware for EXAONE Deep 2.4B

EXAONE Deep 2.4B — 2.4B Parameter Dense LLM

Model Specifications

Parameters
2.4B
Architecture
Dense Transformer
Context Length
32K tokens
Capabilities
reasoning, math, coding
Release Date
2025-03-16
Family
exaone

VRAM Requirements

QuantizationBPWVRAMQuality
Q4_K_M4.892.0 GB94%
Q5_K_S5.572.2 GB96%
Q5_K_M5.72.2 GB96%
Q6_K6.562.5 GB97%
Q8_08.53.0 GB100%
FP16165.3 GB100%

Benchmark Scores

MMLU-PRO25.3
MATH36.8
IFEval79.5
BBH15.9
GPQA2.1
MUSR3.2

How to Run EXAONE Deep 2.4B

Run EXAONE Deep 2.4B locally with Ollama (needs 2.0 GB VRAM at Q4_K_M):

ollama run exaone-deep:2.4b

Compatible GPUs (30)

GPUs that can run EXAONE Deep 2.4B at Q4_K_M quantization:

AMD FireGL V8650(2GB, 111 GB/s)NVIDIA GeForce GTX 285 X2(2GB, 148 GB/s)NVIDIA GeForce GTX 480M(2GB, 77 GB/s)NVIDIA Quadro 5000M(2GB, 77 GB/s)AMD Radeon HD 6950(2GB, 160 GB/s)AMD Radeon HD 6970(2GB, 176 GB/s)Intel Aubrey Isle(2GB, 154 GB/s)AMD Radeon HD 5870 Eyefinity 6(2GB, 154 GB/s)NVIDIA GeForce GTX 485M(2GB, 96 GB/s)NVIDIA GeForce GTX 580M(2GB, 96 GB/s)NVIDIA Quadro 1000M(2GB, 29 GB/s)NVIDIA Quadro 2000M(2GB, 29 GB/s)NVIDIA Quadro 3000M(2GB, 80 GB/s)NVIDIA Quadro 3000M X2(2GB, 80 GB/s)NVIDIA Quadro 4000M(2GB, 80 GB/s)