EXAONE 4.5 33B (Reasoning)
LG AI ResearchProprietary
描述
EXAONE 4.5 is the first open-weight vision-language model from LG AI Research. It is a dense 33B multimodal model (31.7B language + 1.29B vision encoder) built by adding a dedicated visual encoder to the EXAONE 4.0 framework. It supports a 262,144-token context window, hybrid reasoning (thinking on by default), and strong performance on document understanding and Korean contextual reasoning while retaining EXAONE language capabilities. License: EXAONE AI Model License Agreement 1.2 - NC. Knowledge cutoff: December 2024.
發布日期
2026-04-09
參數規模
33.0B
上下文長度
—
支援模態
—
能力雷達圖
80
general
0
coding
90
reasoning
68
science估算
70
agents
80
multimodal
缺少專門科學評測時,Science 由 LLM Stats 科學得分或推理能力估算。
排行榜排名
| 領域 | #排名 | 分數 | 來源 |
|---|---|---|---|
| 多模態榜 | 81 | 49.0 | LS |
基準測試分數 (LLM Stats)
(LLM Stats (zeroeval))Chat
IFEvalGoogle Research (2023)
89.6%自報
Tau2 Airline
56.5%自報
Communication
Tau2 Retail
77.9%自報
Tau2 Telecom
73.0%自報
Language
MMLU-Pro
83.3%自報
Math
AIME 2025
92.9%自報
AIME 2026
92.6%自報
MathVista-Mini
85.0%自報
Multimodal
MMMU
78.7%自報
Reasoning
LiveCodeBench v6
81.4%自報
GPQANYU + Cohere + Anthropic (2023)
80.5%自報
AA-LCR
50.6%自報
Vision
AI2D
89.0%自報
MMStar
74.9%自報
MMMU-Pro
68.6%自報
AA 評測指數
(Artificial Analysis)暫無 AA 評測資料
LLM Stats 分類評分
(LLM Stats (zeroeval))Instruction Following90
Math90
Structured Output90
Language80
Legal80
Multimodal80
Physics80
Reasoning80
Finance80
General80
Healthcare80
Biology80
Chemistry80
Vision80
Chat70
Communication70
Tool Calling70
Long Context50
定價
暫無定價資料
速度
暫無速度資料
供應商價格排行
暫無提供商資料