LG AI Research
What benchmarks does LG AI Research mention most?
Here are the 21 benchmarks LG AI Research mentions most in its launch posts.
Source: Labs’ launch posts, read by Benchmark Almanac
GPQA Diamond MMLU-Pro AIME 2025 LiveCodeBench v5 IFEval KMMLU-Pro MMMLU WMT24++ LiveCodeBench v6 Ko-LongBench (LG) HRM8K-KSM IFBench AA-LCR (Artificial Analysis Long Context Reasoning) Humanity's Last Exam IMO-AnswerBench SWE-bench Verified tau2-bench OpenAI MRCR KoBALT CLIcK KGC-Safety (LG) AIME 2026
Source: Labs’ system cards and technical reports, read by Benchmark Almanac
GPQA Diamond MMLU-Pro AIME 2025 LiveCodeBench v5 MATH-500 AIME 2024 CSAT Math 2025 IFEval KMMLU-Pro MMMLU LiveCodeBench v6 Ko-LongBench (LG) WMT24++ HRM8K-KSM MMLU-Redux HMMT 2025 Multi-IF Berkeley Function Calling Leaderboard (BFCL) IFBench AA-LCR (Artificial Analysis Long Context Reasoning) Humanity's Last Exam IMO-AnswerBench SWE-bench Verified tau2-bench KoBALT CLIcK KGC-Safety (LG) AIME 2026
Source: Labs’ launch posts, system cards and technical reports, read by Benchmark Almanac
Model cards
| Date | Model | Document | Benchmarks |
|---|---|---|---|
| Jul 31, 2026 | K-EXAONE 2.0 | K-EXAONE-2.0-750B-A37B model card (Hugging Face) | 24 |
| Apr 9, 2026 | EXAONE 4.5 33B | EXAONE-4.5-33B model card (Hugging Face) | 30 |
| Dec 31, 2025 | K-EXAONE | K-EXAONE-236B-A23B model card (Hugging Face) | 25 |
| Jul 15, 2025 | EXAONE 4.0 32B | EXAONE-4.0-32B model card (Hugging Face) | 19 |
| Mar 18, 2025 | EXAONE Deep 32B | EXAONE-Deep-32B model card (Hugging Face) | 6 |
Every LG AI Research model
- K-EXAONE 2.0 Jul 2026
- EXAONE 4.5 33B Apr 2026
- K-EXAONE Dec 2025
- EXAONE 4.0 32B Jul 2025
- EXAONE Deep 32B Mar 2025