EXAONE 4.5 33B — Hardware Requirements & GPU Compatibility
Vision
EXAONE 4.5 is LG AI Research's 33-billion-parameter vision-language model, the first open-weight vision-language release from the company. It adds a dedicated 1.29-billion-parameter vision encoder to the EXAONE 4.0 language model (31.7 billion parameters) and uses a hybrid attention design with sliding-window and global layers. The card says it outperforms state-of-the-art models of similar size in document understanding and Korean contextual reasoning, and it supports English, Korean, Spanish, German, Japanese and Vietnamese. Running it locally calls for a 24 GB or larger GPU once quantized, or a unified-memory machine. The model supports a 262,144 token context window. It is released under the EXAONE AI Model License Agreement 1.2 - NC, a non-commercial license, so commercial use is not permitted under its terms. It was published in April 2026 and builds on the earlier EXAONE 4.0 language models.
52.9K downloads
185 likesSpecifications
- Publisher
- LGAI-EXAONE
- Family
- EXAONE
- Parameters
- 34.4B
- Architecture
- Exaone4_5_ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 153,600
- Release Date
- 2026-04-04
- License
- Other
Get Started
HuggingFace
How Much VRAM Does EXAONE 4.5 33B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 15.4 GB | 83.6 GB | 14.60 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 15.9 GB | 84.0 GB | 15.03 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 17.6 GB | 85.8 GB | 16.75 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 18.0 GB | 86.2 GB | 17.18 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 21.4 GB | 89.6 GB | 20.61 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 25.3 GB | 93.5 GB | 24.47 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 29.2 GB | 97.4 GB | 28.34 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 35.2 GB | 103.4 GB | 34.35 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run EXAONE 4.5 33B?
Q4_K_M · 21.4 GBEXAONE 4.5 33B (Q4_K_M) requires 21.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 28+ GB is recommended. Using the full 262K context window can add up to 68.2 GB, bringing total usage to 89.6 GB. 7 GPUs can run it, including NVIDIA GeForce RTX 5090.
All compatible consumer-level GPUs are running near their VRAM limit. You may also want to consider professional GPUs (e.g., NVIDIA A100, H100) which offer significantly more VRAM. For more headroom and better throughput, consider a multi-GPU configuration with tensor parallelism (supported by tools like vLLM, llama.cpp, or text-generation-inference).
Which Devices Can Run EXAONE 4.5 33B?
Q4_K_M · 21.4 GB41 devices with unified memory can run EXAONE 4.5 33B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightWhere to Download EXAONE 4.5 33B
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does EXAONE 4.5 33B need?
EXAONE 4.5 33B requires 21.4 GB of VRAM at Q4_K_M, or 69.5 GB at BF16. Full 262K context adds up to 68.2 GB (89.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 34.4B × 4.8 bits ÷ 8 = 20.6 GB
KV Cache + Overhead ≈ 0.8 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 69 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M21.4 GBQ4_K_M + full context89.6 GB- Can NVIDIA GeForce RTX 4090 run EXAONE 4.5 33B?
Yes, at Q4_K_M (21.4 GB) or lower. Higher quantizations like Q5_K_S (24.4 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for EXAONE 4.5 33B?
For EXAONE 4.5 33B, Q4_K_M (21.4 GB) offers the best balance of quality and VRAM usage. Q5_K_S (24.4 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 15.4 GB.
VRAM requirement by quantization
Q2_K15.4 GBQ3_K_M17.6 GBQ4_120.2 GBQ4_K_M ★21.4 GBQ5_K_S24.4 GBBF1669.5 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run EXAONE 4.5 33B on a Mac?
EXAONE 4.5 33B requires at least 15.4 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run EXAONE 4.5 33B locally?
Yes — EXAONE 4.5 33B can run locally on consumer hardware. At Q4_K_M quantization it needs 21.4 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is EXAONE 4.5 33B?
At Q4_K_M, EXAONE 4.5 33B can reach ~224 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~31 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 21.4 × 0.65 = ~242 tok/s
Estimated speed at Q4_K_M (21.4 GB)
~242 tok/s~31 tok/s~242 tok/s~224 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of EXAONE 4.5 33B?
At Q4_K_M, the download is about 20.61 GB. The full-precision BF16 version is 68.70 GB. The smallest option (Q2_K) is 14.60 GB.
- Which GPUs can run EXAONE 4.5 33B?
7 consumer GPUs can run EXAONE 4.5 33B at Q4_K_M (21.4 GB). Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.
- Which devices can run EXAONE 4.5 33B?
41 devices with unified memory can run EXAONE 4.5 33B at Q4_K_M (21.4 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB), Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.