K-EXAONE 2.0

K-EXAONE 2.0 is a reasoning model from LG AI Research. 18 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.50.3 ±5.3
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.
Input / 1MUS dollars per 1M input tokens.Free
Output / 1MUS dollars per 1M output tokens.Free
ContextMaximum tokens in one request.262K
EloLMArena rating and rank.N/A

The index is a score out of 100. The ± range shows how much it can change.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
65.2
CodingCode writing and repair.
50.2
ReasoningLogic problems and puzzles.
43.7
MultimodalTasks with images and text.
N/A
KnowledgeFacts and expert knowledge.
46.5
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
50.4
MathMath problems.
50.5

Results

18 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
Instruction-Following EvalInstruction92.4%50.4Jeffrey Zhou et al.
AIME 2026Math92.3%52.2Qwen
MMMLUKnowledge86.6%OpenAI
Massive Multitask Language Understanding ProfessionalKnowledge83.5%52.1Yubo Wang et al.
Artificial Analysis GPQA DiamondKnowledge82.9%53.5Artificial Analysis
GPQA DiamondKnowledge82.2%54.1David Rein et al.
IMOAnswerBenchMath78.6%DeepSeek-AI
Harvard-MIT Mathematics Tournament February 2026Math78.4%48.8Qwen
Claw-EvalAgentic77.7%80.5Bowen Ye et al.
Instruction Following BenchmarkInstruction72.6%50.5Benchmark authors
PolyMathMultilingual71.3%Qwen
Software Engineering Benchmark VerifiedCoding68.2%52.2Carlos E. Jimenez et al.
Artificial Analysis Long Context ReasoningReasoning56.2%47.1Artificial Analysis
Terminal-Bench 2.1 (provider run)Agentic43.8%49.9DeepSeek-AI
Terminal-Bench 2.1 (provider run)Agentic43.8%49.9DeepSeek-AI
Artificial Analysis SciCodeCoding42.0%51.2Artificial Analysis
Scientific Code BenchmarkCoding37.4%47.1Benchmark authors
Artificial Analysis Intelligence IndexKnowledge19.7%47.0Artificial Analysis
Artificial Analysis Humanity's Last ExamKnowledge18.6%44.7Artificial Analysis
Humanity's Last ExamKnowledge18.3%44.3Center for AI Safety et al.
Artificial Analysis Omniscience AccuracyKnowledge13.1%30.1Artificial Analysis
Critical Physics TasksReasoning0.9%40.3Artificial Analysis

18 benchmarks count, from 19 of 22 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authors

More from LG AI Research

K-Exaone43.6Exaone 4.0 32B33.2Exaone 4.0 1.2B22.9