EXAONE-4.5

We introduce EXAONE 4.5, the first open-weight vision language model developed by LG AI Research. Integrating a dedicated visual encoder into the existing EXAONE 4.0 framework, we expand the model's capability toward multimodality. EXAONE 4.5 features 33 billion parameters in total, including 1.2 billion parameters from the vision encoder. EXAONE 4.5 achieves competitive performance in general benchmark while outperforming SOTA models of similar size in document understanding and Korean contextual reasoning, inheriting powerful language capabilities from our previous language models.

For more details, please refer to the technical report, blog and GitHub.

Model Configuration

• Model Type: Causal Language Model + Vision Encoder
• Number of Parameters (Language Model): 31.7B
• Number of Parameters (Vision Encoder): 1.29B
• Hidden Dimension: 5,120
• Intermediate size: 27,392
• Number of Layers: 64 Main layers + 1 MTP layers
  • Hybrid Attention Pattern: 16 x (3 Sliding window attention + 1 Global attention)
  • Reordered Norm: Apply normalization after Attention/MLP, and before residual connection
• Sliding Window Attention
  • Number of Attention Heads: 40 Q-heads and 8 KV-heads
  • Head Dimension: 128 for both Q/KV
  • Sliding Window Size: 4096
• Global Attention
  • Number of Attention Heads: 40 Q-heads and 8 KV-heads
  • Head Dimension: 128 for both Q/KV
  • No Rotary Positional Embedding Used (NoPE)
• Vision Encoder
  • Grouped Query Attention (GQA)
  • 2D RoPE for vision embeddings
• Vocab Size: 153,600
• Context Length: 262,144 tokens
• Knowledge Cutoff: Dec 2024 (2024/12)

Evaluation Results

Vision-Language Tasks

Language-only Tasks

Quickstart

Serving EXAONE 4.5

For better inference speed and memory usage, it is preferred to serve the model using optimized inference engines. The EXAONE 4.5 model is supported by various frameworks, including TensorRT-LLM, vLLM, SGLang, and llama.cpp. Support will be expanded in the future.

Practically, you can serve the EXAONE 4.5 model with 256K context length on single H200 GPU, or 4x A100-40GB GPUs by using a tensor-parallelism.

TensorRT-LLM

TensorRT-LLM provides zero day support for EXAONE 4.5. Please refer to the official installation guide, and EXAONE documentations.

You should also install the Transformers library with transformers >= 5.8.0.

After installing TensorRT-LLM and Transformers, you can launch the server with the following code snippet. You may remove any unnecessary arguments from the snippet.

trtllm-serve LGAI-EXAONE/EXAONE-4.5-33B \
    —tp_size 2 \
    —port 8000 \
    —reasoning_parser qwen3

An OpenAI-compatible API server will be available at http://localhost:8000/v1.

vLLM

The EXAONE 4.5 model is officially supported in transformers >= 5.8.0 and vllm >= 0.5.8.

After installing vLLM, you can launch the server with the following code snippet. You can remove unnecessary arguments from the snippet.

vllm serve LGAI-EXAONE/EXAONE-4.5-33B \
    --served-model-name EXAONE-4.5-33B \
    --port 8000 \
    --tensor-parallel-size 2 \
    --max-model-len 262144 \
    --reasoning-parser qwen3 \
    --enable-auto-tool-choice \
    --tool-call-parser hermes \
    --limit-mm-per-prompt '{"image": 64}' \
    --speculative_config '{
        "method": "mtp",
        "num_speculative_tokens": 3
    }'

An OpenAI-compatible API server will be available at http://localhost:8000/v1.

SGLang

To run the EXAONE 4.5 model with SGLang, you need the Transformers library and our fork of SGLang.

For Transformers, install transformers >= 5.8.0. For SGLang, install the required dependencies by running the following command:

uv pip install 'git+https://github.com/lkm2835/sglang.git@add-exaone4_5#subdirectory=python&egg=sglang[all]'

After installing the SGLang and transformers, you can launch the server with the following code snippet. You can remove unnecessary arguments from the snippet.

python -m sglang.launch_server \
    --model-path LGAI-EXAONE/EXAONE-4.5-33B \
    --served-model-name EXAONE-4.5-33B \
    --port 8000 \
    --tp-size 2 \
    --mem-frac 0.81 \
    --reasoning-parser qwen3 \
    --tool-call-parser hermes \
    --speculative-algorithm EAGLE \
    --speculative-num-steps 3 \
    --speculative-eagle-topk 1 \
    --speculative-num-draft-tokens 4

An OpenAI-compatible API server will be available at http://localhost:8000/v1.

Using EXAONE 4.5

After launching the OpenAI-compatible server with EXAONE 4.5, you can seamlessly use the model via API with a single code integration, even though the serving framework has changed. To use OpenAI Python SDK and following examples, you should install the openai library on your environment.

[!IMPORTANT] To achieve the expected performance, we recommend using the following configurations:

• We recommend to use temperature=1.0, top_p=0.95, presence_penalty=1.5 for general purpose.
• We recommend to use temperature=0.6, top_p=0.95, presence_penalty=1.5, top_k=20 for OCR/document-related tasks, and Korean inputs.
• We recommend to use temperature=1.0, top_p=0.95 for text-only inputs.
• Different from EXAONE-4.0, EXAONE 4.5 uses enable_thinking=True as default. Thus, you need to set enable_thinking=False when you want to use non-reasoning mode.
• EXAONE 4.5 prefers using \boxed{} format to answer the question. We recommend using this format with the corresponding format instruction for better parsing accuracy.

You can easily try model's chat completions by using OpenAI Python SDK. For your server in local machine, you will need to change your base_url and api_key f