EXAONE 4.5 33B
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M LGAI-EXAONE/EXAONE-4.5-33B-GGUF | 18.67 GiB | 1 | 20.9 GiB | 24 GB GPU | HF files |
| Q5_K_M LGAI-EXAONE/EXAONE-4.5-33B-GGUF | 21.87 GiB | 1 | 24.1 GiB | 32 GB GPU | HF files |
| Q8_0 LGAI-EXAONE/EXAONE-4.5-33B-GGUF | 32.73 GiB | 1 | 35.0 GiB | 48 GB GPU | HF files |
No verified file: FP16, MLX 4-bit, MLX 8-bit. Sizes are exact file bytes (text weights; projectors listed separately). “+” = recurrent state or other terms unknown. Smallest fit adds 0.5 GiB spare; Macs use 75% of memory.
Quality
Artificial Analysis Intelligence Index
13.2 AA estimate
Variant: EXAONE 4.5 33B (Reasoning) · observed 2026-10-17 · source
Rank #46 of 75 researched 2026 models.
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 0
No reported measurement yet. Submit one.
Community threads
- vLLM (add-exaone4_5 branch) 빌드 실패 관련 문의드립니다. Hugging Face · 2026-05-04
- EXAONE 4.5 33B FP8 vLLM ROCm 구동 중 ImageProcessor / ProcessorMixin 정합성 이슈 Hugging Face · 2026-04-29
Qualitative signals only; titles are not verified claims.