LG AI ResearchEXAONE 4.5 33B
EXAONE 4.5 33B is a LG AI Research open-weights reasoning model, released 9 Apr 2026. Its best configuration (thinking reasoning effort) ranks #420 of 785 on the BenchLeader Index at 48.6 ±4.7, in the lower half. The ± is the point: 327 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (58) and lowest in knowledge (39). It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 9 Apr 2026.stale: no new result in six months
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- 262k
- Full answer
- –
- Cost per run
- –
- Released
- 9 Apr 2026
- Overall index49
- Reasoning41
- Agents & tools52
- Knowledge39
- Instruction following58
- Multimodal50
- Long context52
- Composite44
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. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | – | – | – | |
| thinkingbest | 48.6 | #420 | – | – | – | Agents & tools 52 · Composite 44 · Instruction following 58 · Knowledge 39 · Long context 52 · Multimodal 50 · Reasoning 41 |
How its index has moved
2 Oct 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. 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
| Benchmark | thinking | Source |
|---|---|---|
| GPQA Diamond (AA)not in index | 79.4%#198 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 12.9%#277 | Artificial Analysis |
| CritPt | 0.3%#271 | Artificial Analysis |
Coding
| Benchmark | thinking | Source |
|---|---|---|
| Terminal-Bench 2.1 (AA)not in index | 21.4%#170 | Artificial Analysis |
Agents & tools
| Benchmark | thinking | Source |
|---|---|---|
| Terminal-Bench Hard | 20.4%#163 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 78.1%#131 | Artificial Analysis |
| GDPval-AA v2.1not in index | 0.0%#124 | Artificial Analysis |
Knowledge
| Benchmark | thinking | Source |
|---|---|---|
| AA-Omniscience | -50.7#391 | Artificial Analysis |
| AA-Omniscience: accuracynot in index | 16.2%#396 | Artificial Analysis |
| AA-Omniscience: non-hallucinationnot in index | 20.1%#276 | Artificial Analysis |
Instruction following
| Benchmark | thinking | Source |
|---|---|---|
| IFBench | 58.0%#134 | Artificial Analysis |
Multimodal
| Benchmark | thinking | Source |
|---|---|---|
| MMMU-Pro | 67.3%#167 | Artificial Analysis |
Long context
| Benchmark | thinking | Source |
|---|---|---|
| AA-LCR | 54.7%#302 | Artificial Analysis |
Composite
| Benchmark | thinking | Source |
|---|---|---|
| AA Intelligence Index v4.3.2 | 13.2#313 | Artificial Analysis |
More from LG AI Research
See also
Questions people ask
- What are EXAONE 4.5 33B's benchmark scores?
- As of 25 Oct 2026, EXAONE 4.5 33B (thinking effort) scores 48.6 ±4.7 on the BenchLeader Index, strongest in instruction following (58) and weakest in knowledge (39). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does EXAONE 4.5 33B rank?
- EXAONE 4.5 33B (thinking effort) is #420 of 785 ranked configurations on 25 Oct 2026. 327 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- When was EXAONE 4.5 33B released?
- LG AI Research released EXAONE 4.5 33B on 9 Apr 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 25 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “EXAONE 4.5 33B: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/exaone-4-5-33b.