VerdictPal
Join

Models / Other

EXAONE 4.5 33B

EXAONE 4.5 · Released 2026-04-09

AA Intelligence Index 20.2. Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#113 of 131 on AA Index · current snapshot
20 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite35
ContextUnlisted
Input / 1MFree
Output / 1MFree
Knowledge cutoffunpublished
Output speed0 t/s
TTFT (default API)0.00s
Statussolid
ReasoningCodingAgenticLong context
Reasoning
79
Coding
28
Agentic
21
Long context
49

44average across 4 tested families

Benchmark placements

Where EXAONE 4.5 33B 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
IFBenchReasoning#81/ 12158Artificial Analysis2026-07-28
AA output speedPerformance#87/ 149Artificial Analysis2026-07-28
AA time to first tokenPerformance#87/ 149Artificial Analysis2026-07-28
GPQA DiamondReasoning#112/ 14979Artificial Analysis2026-07-28
AA-LCRLong context#117/ 14349Artificial Analysis2026-07-28
Humanity's Last ExamReasoning#118/ 14412Artificial Analysis2026-07-28
τ³-Bench BankingAgentic#119/ 14311Artificial Analysis2026-07-28
Artificial Analysis Intelligence IndexReasoning#127/ 14720Artificial Analysis2026-07-28
SciCodeCoding#132/ 14428Artificial Analysis2026-07-28
Terminal-BenchAgentic#132/ 14821Artificial Analysis2026-07-28
Family context

The three highest-scoring carded models in each capability family. Where EXAONE 4.5 33B shows up, it's highlighted.

ReasoningEXAONE 4.5 33B · 79
  1. 1
    Claude Mythos Preview
    95
  2. 2
    GPT-5.6 Sol
    94
  3. 3
    Kimi K3
    94
CodingEXAONE 4.5 33B · 28
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticEXAONE 4.5 33B · 21
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextEXAONE 4.5 33B · 49
  1. 1
    GPT-5 (high)
    76
  2. 2
    GPT-5.2 Codex (xhigh)
    76
  3. 3
    Kimi K3
    75
Newest receipts

Benchmark rows added to the public ledger for EXAONE 4.5 33B in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context49.33%Artificial Analysis4
AA output speedPerformance0 t/sArtificial Analysis4
AA time to first tokenPerformance0.00sArtificial Analysis4
Artificial Analysis Intelligence IndexReasoning20.2%Artificial Analysis4
GPQA DiamondReasoning79.4%Artificial Analysis4
Humanity's Last ExamReasoning11.6%Artificial Analysis4
IFBenchReasoning57.96%Artificial Analysis4
SciCodeCoding28%Artificial Analysis4
τ³-Bench BankingAgentic11.13%Artificial Analysis4
Terminal-BenchAgentic21.35%Artificial Analysis4
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.

Reasoning79
Coding28
Agentic21
Long context49
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.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotxAIMiniMaxMetaZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for EXAONE 4.5 33B. Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 32 catalog benchmarks have a sourced row for EXAONE 4.5 33B.

Reasoning

Coding

Agentic

Long context

Performance

Sources