LG AI Research model product
EXAONE 4.5 33B
5 33B is an open-weight vision-language model from LG AI Research with a dense 33B multimodal architecture.
Updated Sep 8, 2026. Default version: EXAONE 4.5 33B
LLMBoard Score44.1EXAONE 4.5 33B
Coverage0%15 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable
EXAONE 4.5 33B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
EXAONE 4.5 33B Benchmark Results
Benchmark scores for EXAONE 4.5 33B.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkMathVista-Mini | Score85.00% | Rank09 | Participants25 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkTau2 Retail | Score77.90% | Rank10 | Participants27 | Percentile65.38% | EvidenceC | Evaluated |
| BenchmarkMMStar | Score74.90% | Rank12 | Participants25 | Percentile54.17% | EvidenceC | Evaluated |
| BenchmarkAIME 2026 | Score92.60% | Rank13 | Participants22 | Percentile42.86% | EvidenceC | Evaluated |
| BenchmarkAI2D | Score89.00% | Rank15 | Participants34 | Percentile57.58% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score78.70% | Rank16 | Participants64 | Percentile76.19% | EvidenceC | Evaluated |
| BenchmarkTau2 Airline | Score56.50% | Rank16 | Participants24 | Percentile34.78% | EvidenceC | Evaluated |
| BenchmarkAA-LCR | Score50.60% | Rank17 | Participants19 | Percentile11.11% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score89.60% | Rank21 | Participants68 | Percentile70.15% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v6 | Score81.40% | Rank21 | Participants62 | Percentile67.21% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score73.00% | Rank25 | Participants36 | Percentile31.43% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score92.90% | Rank33 | Participants119 | Percentile72.88% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score83.30% | Rank34 | Participants138 | Percentile75.91% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score68.60% | Rank40 | Participants69 | Percentile42.65% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score80.50% | Rank94 | Participants247 | Percentile62.20% | EvidenceC | Evaluated |
EXAONE 4.5 33B Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
EXAONE 4.5 33B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
EXAONE 4.5 33B Runtime Performance
Provider-specific output speed and catalog latency for EXAONE 4.5 33B. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
EXAONE 4.5 33B Specifications
Technical details for the model's default version.
- Version
- EXAONE 4.5 33B
- Released
- Apr 9, 2026
- Knowledge cutoff
- Dec 1, 2024
- Parameters
- 33B
- Context window
- N/A
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Proprietary
EXAONE 4.5 33B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionEXAONE 4.5 33B | Released | LLMBoard44.11 | Parameters33B | ContextN/A | Max outputN/A | Open weightsYes | LicenseProprietary |
Models similar to EXAONE 4.5 33B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
LA
K EXAONE 236B A23B
LG AI Research
OP
GPT-OSS-120B
OpenAI
OP
GPT-5-Codex
OpenAI
NV
Nemotron 3 Super
NVIDIA
OP
o4 mini
OpenAI
AC
Qwen3 VL 235B A22B Thinking
Alibaba Cloud / Qwen Team
What is EXAONE 4.5 33B?
Key information about EXAONE 4.5 33B and its available data.
0 framework. It supports a 262,144-token context window and hybrid reasoning, with thinking enabled by default.
Data as of 2026-09-08.
FAQ
Common questions about EXAONE 4.5 33B.
When was EXAONE 4.5 33B released?
EXAONE 4.5 33B's default version was released on Apr 9, 2026.
How much does EXAONE 4.5 33B cost?
No official standard PAYG price is currently available for EXAONE 4.5 33B.
Who created EXAONE 4.5 33B?
EXAONE 4.5 33B was created by LG AI Research.
What is the context window for EXAONE 4.5 33B?
A context window is not available for the default version.
Is EXAONE 4.5 33B open weight?
Yes. The default version is marked as open weight under Proprietary.
How many API providers offer EXAONE 4.5 33B?
No provider offering is currently linked to the default version.