UnifyBench

RankingEffort benchmarks

K EXAONE 236B A23B effort benchmarks

Compare measured Low, Medium, High and other settings. Every number keeps its source.

K EXAONE 236B A23B

27 measured results · 2 reported effort labels · 15 benchmark/harness combinations. Model evidence and sources →

Capability score evidence

K EXAONE 236B A23B · Reasoning: 8.9 overall comparison score. Meets the overall evidence requirements.

Which sources carry the weight?
  • Artificial Analysis · 100.0% direct comparison weight · independent

These shares describe the comparisons entering the fit, not fractions of the final score.

How sensitive is this score?

Remove one evaluator or family and refit the graph against the same overall panel. Values use the matching dataset audited 2026-09-12. These are sensitivity checks, not confidence intervals; ranking eligibility may also change.

Removed evidenceRecomputed scoreChange
Artificial AnalysisComparison path lost—
ARC Prize10.5+1.7
Datacurve7.6-1.3
LiveCodeBench8.8-0.1
Scale9.8+0.9
Terminal-Bench8.9-0.0
Cognition9.3+0.4
LiveBench9.1+0.2
SWE-rebench9.1+0.2
Epoch AI8.9+0.0
Vals AI8.6-0.2
Provider reports · Anthropic8.9+0.0
Provider reports · Moonshot AI8.9-0.0
Provider reports · Meta8.9-0.0
Provider reports · Mistral8.9+0.0
All provider reports8.9-0.0
Every benchmark family
Removed evidenceRecomputed scoreChange
arc-agi10.5+1.6
deepswe7.6-1.3
gmmlu9.6+0.8
gdpval9.5+0.7
mmmu-pro9.4+0.5
ifbench8.3-0.5
epoch-game-puzzles8.4-0.4
frontiercode9.3+0.4
omniscience9.3+0.4
critpt8.5-0.4
hle9.3+0.4
aime8.5-0.4
livecodebench8.5-0.3
simpleqa9.2+0.3
swe-rebench9.1+0.2
gpqa8.7-0.2
tau-bench8.7-0.2
multi-swe-bench8.7-0.1
livebench-instructions9.0+0.1
scicode9.0+0.1
harvey8.8-0.1
terminal-bench9.0+0.1
aa-lcr8.8-0.1
livebench-language9.0+0.1
frontiermath9.0+0.1
swe-atlas-qna8.8-0.1
apex-agents9.0+0.1
gdp-pdf8.9+0.1
livebench-coding8.8-0.1
arc-agi-38.8-0.1
vibe-code8.8-0.1
livebench-reasoning8.8-0.1
briefcase8.8-0.0
vals-finance-agent8.8-0.0
math5008.9+0.0
automationbench8.8-0.0
vals-legal-research8.8-0.0
swe-atlas-test-writing8.8-0.0
vals-code-migration8.8-0.0
itbench8.9-0.0
analyst-agent8.9+0.0
livebench-data8.9+0.0
livebench-math8.9+0.0
vals-excel-modeling8.9-0.0
swe-atlas-refactoring8.9-0.0
enterprise-ops8.9-0.0
enigma-eval8.9+0.0
mmlu-pro8.9+0.0
swe-bench-pro8.9+0.0
What evidence is missing from the fit?

Across all collected settings for this model: 27 observations, 25 contributing, 0 matched without graph weight and 2 excluded. Counts do not establish rank eligibility.

  • 2 · Benchmark family or source protocol has not been reviewed
Inspect every observation and exclusion →

Capability profile

One configuration, seven capabilities

Comparison score · 0–100
K EXAONE 236B A23B · Reasoning capability radarComparison scores from 0 to 100. Unknown capabilities have no point; lines stop at gaps. Hollow points are preliminary. Exact values and support labels follow the chart.AgenticHard reasoningCodingHuman prefKnowledgeMultimodalLong context50100K EXAONE 236B A23B · Reasoning · Agentic: 8.9 · SupportedK EXAONE 236B A23B · Reasoning · Hard reasoning: 14.1 · SupportedK EXAONE 236B A23B · Reasoning · Coding: 10.4 · SupportedK EXAONE 236B A23B · Reasoning · Knowledge: 8.9 · SupportedK EXAONE 236B A23B · Reasoning · Long context: 13.6 · Preliminary

Gaps are unknown, not zero. Hollow points are preliminary.

  • Agentic

    A8.9Supported

  • Hard reasoning

    A14.1Supported

  • Coding

    A10.4Supported

  • Human pref

    Unknown

  • Knowledge

    A8.9Supported

  • Multimodal

    Unknown

  • Long context

    A13.6Preliminary

Each profile belongs to the named configuration. Capability scores use a shared panel for all axes, distinct from the overall ranking panel; they are not accuracy percentages or direct head-to-head wins.

Compare this configuration →Explore the model evidence profile →

These are published benchmark scores. Supported effort estimates also appear on the Capability leaderboard. Results from different harnesses occupy different rows. Missing cells stay unknown; no scores are copied between effort levels.

Benchmark / sourceReasoningNon-reasoning
AIME 2025Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

90.33%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

44%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

AA-LCR v1.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

61.33%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

53.33%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

CritPtArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by Argonne and UIUC, with contributions from 60+ researchers globally · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

1.12%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

0%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

GDPval-AA v2 LeaderboardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Elo rating for performance on real-world work tasks · Anchored to a human baseline of 1,000 · Higher is better · Evaluation results measured independently by Artificial Analysis

Unit: elo. Source collected 2026-09-12.

539.71 Elo
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

Not reported
Global-MMLU-LiteArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

78.86%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

71.03%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

GPQA DiamondArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

78.28%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

69.49%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

Humanity's Last ExamArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

13.95%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

5.7%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

IFBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

64.69%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

39.59%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

LiveCodeBenchArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

76.83%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

Not reported
MMLU-ProArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

83.75%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

80.98%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

AA-Omniscience IndexArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

Unit: index. Source collected 2026-09-12.

-57.97
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

-65.4
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

AA-Omniscience AccuracyArtificial Analysis · 6000 questions; 1 attempt; correct-answer share; GPT-5.6 Luna medium grader
Benchmark details

AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

16.35%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

13.65%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

𝜏²-Bench TelecomArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

74.27%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

59.06%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

Terminal-Bench HardArtificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

22.73%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

6.82%
Reported settings

K-EXAONE (Non-reasoning); Model source: https://artificialanalysis.ai/models/k-exaone-non-reasoning

Terminal-Bench v2.1Artificial Analysis · Artificial Analysis published evaluation; see source for benchmark-specific settings
Benchmark details

Benchmark developed by the Laude Institute and open-source contributors · Independently benchmarked by Artificial Analysis · Evaluation results measured independently by Artificial Analysis

Unit: percent. Source collected 2026-09-12.

30.34%
Reported settings

K-EXAONE (Reasoning); Model source: https://artificialanalysis.ai/models/k-exaone

Not reported
What these comparisons mean

Each source reports its own effort labels. Matching labels do not establish equal compute budgets or identical fallback behavior. Open Reported settings to inspect the exact model name, agent and task coverage. Multiple results at one level remain visible rather than selecting the best. Source-reported None or Non-reasoning labels are not treated as evidence that an API supports those settings.

These measurements use their own dated source collection. The leaderboard fits identified configurations jointly; it contains no pooled model ratings. Explicit thinking budgets are separate configurations. Read the methodology.

Download all effort benchmark results · 7,424 results across 240 models · Collected 2026-09-12