# K-Exaone Benchmark Scores & Performance

> K-Exaone by LG AI Research scores 43.48/100 overall, ranking #157 out of 503 AI models.

Canonical page: https://benchlm.ai/models/k-exaone

Last updated: September 22, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | LG AI Research |
| Source Type | Proprietary |
| Reasoning Type | Reasoning |
| Context Window | 256K |
| Overall Score | 43.48/100 |
| Overall Rank | #157 of 503 |

## Family & Coverage

- Family: K-Exaone
- Variant: base
- Benchmarks covered: 13 of 447
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [τ²-bench results](/benchmarks/tau2-bench) | 74.3% |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 0.0% |
| [GDPval-AA](/benchmarks/gdpvalaa) | 540 |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA Coding Index](/benchmarks/aacodingindex) | 32.1% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 61.3% |
| [CritPt](/benchmarks/critpt) | 1.1% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 14.4% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 78.3% |
| [AA-HLE](/benchmarks/aahle) | 13.9% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | -58.0% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 16.4% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 88.8% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-IFBench](/benchmarks/aaifbench) | 64.7% |

## Other LG AI Research Models

- [Exaone 4.0 32B](/models/exaone-4-0-32b) - Score: 39.23
- [Exaone 4.0 1.2B](/models/exaone-4-0-1-2b) - Score: 38.65
- [K-EXAONE 2.0](/models/k-exaone-2-0) - Score: not computed