K-EXAONE
K-EXAONE is behind the leaders in long context, instruction following, reasoning, coding, and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, multilingual tasks, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Safety, Math, Multimodal, Multilingual or Factuality.
Price
No current price is tracked for this model. See the rate card
Evidence
32results on17benchmarks
- 32 aggregator
From 1 source · latest Oct 8, 2026 · How verification works
Research
1 paper reference K-EXAONEK-EXAONE benchmark results
32 results on 17 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
31.0% behind the leader1 of 3 ranked benchmarks measured
- 61.33Oct 8, 2026
Show 1 more long context resultHide 1 long context result
- 53.33Oct 8, 2026
31.8% behind the leader1 of 3 ranked benchmarks measured
- IFBench64.69Oct 8, 2026aa_ifbench
Show 1 more instruction following resultHide 1 instruction following result
- IFBench39.59Oct 8, 2026aa_ifbench
37.7% behind the leader3 of 6 ranked benchmarks measured
- GPQA Diamond78.28Oct 8, 2026gpqa
- Humanity's Last Exam13.95Oct 8, 2026aa_hle
- 1.12Oct 8, 2026
Show 3 more reasoning resultsHide 3 reasoning results
- 0.00Oct 8, 2026
- GPQA Diamond69.49Oct 8, 2026gpqa
- Humanity's Last Exam5.70Oct 8, 2026aa_hle
38.5% behind the leader3 of 10 ranked benchmarks measured
- SciCode35.65Sep 4, 2026aa_scicode
- Terminal-Bench Hard22.73Oct 8, 2026aa_terminalbench_hard
- Terminal-Bench 2.130.34Oct 8, 2026terminalbenchV21
Show 2 more coding resultsHide 2 coding results
- SciCode26.97Sep 4, 2026aa_scicode
- Terminal-Bench Hard6.82Oct 8, 2026aa_terminalbench_hard
46.3% behind the leader2 of 7 ranked benchmarks measured
- τ-Bench V3 · Banking14.23Aug 10, 2026tauBanking
- 0.00Oct 8, 2026
Show 1 more agentic resultHide 1 agentic result
- 13.24Jun 15, 2026
0 of 4 ranked benchmarks measured
- AA-Omniscience · Accuracy13.65Oct 8, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination8.45Oct 8, 2026omniscienceNonHallucination
- AA-Omniscience · Accuracy16.35Oct 8, 2026omniscienceAccuracy
Show 1 more factuality resultHide 1 factuality result
- AA-Omniscience · Non-hallucination11.16Oct 8, 2026omniscienceNonHallucination
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- τ²-Bench Telecom (AA run)59.06Oct 8, 2026aa_tau2
- AA Intelligence11.21Oct 8, 2026aa_intelligence_index
- -65.40Oct 8, 2026
- AA Intelligence14.37Oct 8, 2026aa_intelligence_index
- -57.97Oct 8, 2026
- τ²-Bench Telecom (AA run)74.27Oct 8, 2026aa_tau2
Show 4 more resultsHide 4 results
- 8.00Aug 10, 2026
- 31.22Jun 18, 2026
- Artificial Analysis Coding Index32.11Sep 9, 2026aa_coding_index
- 13.53Jun 18, 2026
K-EXAONE: common questions
Who makes K-EXAONE?
K-EXAONE is made by LG AI.
When was K-EXAONE released?
K-EXAONE was released on Dec 31, 2025, according to Artificial Analysis.
What is K-EXAONE good at?
K-EXAONE is behind the leaders in long context, instruction following, reasoning, coding, and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, multilingual tasks, or factuality.
How many benchmarks has K-EXAONE been tested on?
We track 32 results for K-EXAONE on 17 benchmarks from 1 source. The latest was recorded on Oct 8, 2026.
About this record
Where K-EXAONE's numbers come from, and every name it appears under.
- Tracked since
- May 2, 2026
- Newest source mention
- Jul 20, 2026
Where the results come from
Verification: 32 scores · 0 independently verified · 32 aggregator-attributed. How these tiers are assigned
From 1 source on 1 site. Artificial Analysis supplies all of them. Bars are coloured by trust tier.
- artificialanalysis.ai32