BenchLeader

EXAONE 4.0 32B

Not yet ranked: too few independent results so far.

Blended price
Output speed
First answer
Context
131k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Agents & tools37
  2. Knowledge36
  3. Instruction following39
  4. Composite39

Reasoning-effort configurations

The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.

EffortIndexRankSpeedFirst answerChat reply costCategories
thinkingbestAgents & tools 37 · Composite 39 · Instruction following 39 · Knowledge 36
defaultAgents & tools 35 · Composite 37 · Instruction following 37 · Knowledge 34

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkthinkingdefaultSource
GPQA Diamond (AA)not in index73.9%#25262.8%#342Artificial Analysis
Humanity's Last Exam (AA)not in index11.4%#2295.0%#377Artificial Analysis

Agents & tools

BenchmarkthinkingdefaultSource
Terminal-Bench Hard3.8%#2831.5%#321Artificial Analysis
τ²-Bench Telecom (AA)not in index17.3%#3364.1%#374Artificial Analysis

Knowledge

BenchmarkthinkingdefaultSource
AA-Omniscience-60.1#390-62.8#408Artificial Analysis

Instruction following

BenchmarkthinkingdefaultSource
IFBench36.3%#28733.5%#312Artificial Analysis

Composite

BenchmarkthinkingdefaultSource
AA Intelligence Index8.2#3666.3#450Artificial Analysis

See also

Data as of 9 Sept 2026. Compare these configurations.