EXAONE 4.5 33B
MultimodalEXAONE 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.
Key Specifications
Timeline
Technical Specifications
Benchmark Results
Model performance metrics across various tests and benchmarks
Reasoning
Multimodal
Other Tests
License & Metadata
Similar Models
All Models
Qwen2.5 VL 32B Instruct
Alibaba
Gemma 3 27B
Gemma 3 12B
DiffusionGemma 26B-A4B
Magistral Medium
Mistral AI
Llama 3.2 90B Instruct
Meta
Mistral Small 3.2 24B Instruct
Mistral AI
Mistral Small 3 24B Base
Mistral AI
Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.