EXAONE 4.5 33B
LGAI-EXAONEothertextvision

EXAONE 4.5 33B

Updated Jun 7, 2026
Parameters
34.4B
Context
262,144
License
other
Updated
Jun 7, 2026

Intelligence benchmarks

Artificial Analysis indexes - compared with the best open and proprietary models

Intelligence

30.2

AA Index

Coding

23.0

AA Index

Agentic

36.5

AA Index

Intelligence Index - EXAONE 4.5 33B vs. the field

Best open-weight models (you can run locally) and leading proprietary models for context.

Claude Opus 4.8 (max)
61.4
closed
GPT-5.5 (xhigh)
60.2
closed
Claude Opus 4.7 (max)
57.3
closed
Gemini 3.1 Pro Preview
57.2
closed
Qwen3.7 Max
56.6
closed
Kimi K2.6
53.9
open
MiMo-V2.5-Pro
53.8
open
EXAONE 4.5 33B
30.2
open

Coding Index comparison

GPT-5.5 (xhigh)
59.1
closed
Claude Opus 4.8 (max)
56.7
closed
Gemini 3.1 Pro Preview
55.5
closed
Claude Opus 4.7 (Non-reasoning, high)
53.1
closed
GPT-5.3 Codex (xhigh)
53.1
closed
DeepSeek V4 Pro (Max)
47.5
open
Kimi K2.6
47.1
open
EXAONE 4.5 33B
23
open

Agentic Index comparison

Claude Opus 4.8 (max)
77.8
closed
GPT-5.5 (xhigh)
74.1
closed
Claude Opus 4.7 (max)
71.3
closed
Gemini 3.5 Flash (medium)
70.4
closed
MiniMax-M3
68.6
closed
MiMo-V2.5-Pro
67.4
open
DeepSeek V4 Pro (Max)
67.2
open
EXAONE 4.5 33B
36.5
open

Benchmark data from Artificial Analysis · updated 2026-06-07.

Standard benchmarks

Performance across standard evaluations

BenchmarkScore
GPQA79.4

Will it run on your hardware?

Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest

Runs on your 24 GB - best at Q4_K_M
1 of 4 quantizations fit EXAONE 4.5 33B with real runtime overhead.
fits tight too big

Need an exact figure for your context length? Use the VRAM calculator.

Run it locally

Copy-paste - running in under a minute

vLLMOpenAI-compatible API
vllm serve LGAI-EXAONE/EXAONE-4.5-33B

New to this? Start with Ollama · serve to many users with vLLM.

Deep dive

Notes, sources, and the full write-up

EXAONE 4.5 33B is a 34.4B-parameter other model from LG AI Research. It scores 30.2 on the Artificial Analysis Intelligence Index (coding 23). At Q4_K_M it needs roughly 20 GB of VRAM, placing it in the 12–24 GB GPU hardware tier.

Benchmarks

Artificial Analysis Intelligence Index - EXAONE 4.5 33B vs. leading closed models:

ModelIntelligenceCodingGPQA
EXAONE 4.5 33B30.22379.4
GPT-5.5 (xhigh)60.259.193.5
Claude Opus 4.8 (max)61.456.792
Gemini 3.1 Pro Preview57.255.594.1
Grok 4.3 (high)53.24190.1

Source: Artificial Analysis (2026-06-07).

Popularity

EXAONE 4.5 33B has 49,340 downloads in the last month on HuggingFace and 164 likes.

Frequently asked

Quick answers to common questions

How much VRAM does EXAONE 4.5 33B need?

EXAONE 4.5 33B with 34.4B parameters needs approximately 20 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.

Is EXAONE 4.5 33B better than other LGAI-EXAONE models?

EXAONE 4.5 33B has 34.4B parameters with 262,144 context - a strong choice for general use.

What license is EXAONE 4.5 33B under?

EXAONE 4.5 33B is released under the other license, making it suitable for most commercial and personal projects.

What hardware runs EXAONE 4.5 33B well?

With 34.4B parameters, EXAONE 4.5 33B requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.

What is the best quantization for EXAONE 4.5 33B?

Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~24 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.

How long can EXAONE 4.5 33B's context window handle?

EXAONE 4.5 33B supports a 262,144-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.

What models compete with EXAONE 4.5 33B?

EXAONE 4.5 33B competes with other 17B–52B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.

Nearby options

Similar models and compatible hardware by spec

Comments coming soon

Configure NEXT_PUBLIC_GISCUS_REPO_ID and NEXT_PUBLIC_GISCUS_CATEGORY_ID at giscus.app to enable.