VerdictPal
Join

Models / Other

K-EXAONE 2.0 0803

EXAONE · Released 2026-08-12

LG AI Research K-EXAONE 2.0 (750B-class Korean sovereign model, 31 Jul 2026 weights). Artificial Analysis Intelligence Index 31.0. AA lists $0 / $0 per 1M tokens.

Open Korean foundation-model lane. Prefer sourced ledger rows over ministry launch copy.

#126 of 166 on AA Index · current snapshot
31 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite41
ContextUnlisted
Input / 1MFree
Output / 1MFree
Knowledge cutoffunpublished
Statussolid

Family profile

Best published score in each covered benchmark family.

56/100 avg
Reasoning
83
Coding
41
Agentic
40
Long context
58

4 tested benchmark families

Editor's note
Recorded from the 2026-08-15 AA snapshot (slug k-exaone-2-0-0803, index 31.0) plus LG AI Research's 31 Jul 2026 note. Editorial depth still needs a desk pass.
Benchmark placements

Where K-EXAONE 2.0 0803 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
τ³-Bench BankingAgentic#87/ 10612Artificial Analysis2026-09-01
Terminal-BenchAgentic#114/ 18440Artificial Analysis2026-09-01
SciCodeCoding#115/ 18041Artificial Analysis2026-09-01
GPQA DiamondReasoning#120/ 18583Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#127/ 18019Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#141/ 18431Artificial Analysis2026-09-01
AA-LCRLong context#144/ 17958Artificial Analysis2026-09-01
AA output speedPerformance#155/ 185Artificial Analysis2026-09-01
AA time to first tokenPerformance#155/ 185Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where K-EXAONE 2.0 0803 shows up, it's highlighted.

ReasoningK-EXAONE 2.0 0803 · 83
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingK-EXAONE 2.0 0803 · 41
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticK-EXAONE 2.0 0803 · 40
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextK-EXAONE 2.0 0803 · 58
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for K-EXAONE 2.0 0803 in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context57.67%Artificial Analysis2
AA output speedPerformance0 t/sArtificial Analysis2
AA time to first tokenPerformance0.00sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning31%Artificial Analysis2
GPQA DiamondReasoning82.9%Artificial Analysis2
Humanity's Last ExamReasoning18.6%Artificial Analysis2
SciCodeCoding41%Artificial Analysis2
τ³-Bench BankingAgentic11.55%Artificial Analysis2
Terminal-BenchAgentic40.45%Artificial Analysis2
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning83
Coding41
Agentic40
Long context58
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for K-EXAONE 2.0 0803. Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for K-EXAONE 2.0 0803.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. K-EXAONE 2.0 0803 appears on Artificial Analysis (index 31.0, $0 / $0).
Sources