EXAONE-4.5-33B
Official · BF16 · safetensors
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|
EXAONE · Vision-language · EXAONE 4.5
EXAONE 4.5 33B: Specs, GPU Requirements and License
EXAONE 4.5 33B is a current Vision language release in the EXAONE family, certified from its immutable official source revision.
Total parameters33.0B
Active parametersDense model
Context window262,144 tokens
Release statusCurrent · Current
ArchitectureExaone4_5_ForConditionalGeneration
ModalitiesMultimodal
Licenseother
Deployment stateReference only
Commercial stateUnknown / Not Recommendable
Release date2026-04-04
Architecture, strengths and use cases
EXAONE 4.5 33B is an official self-hostable LG AI Research release in the EXAONE Vision language track. It uses Exaone4_5_ForConditionalGeneration, carries 33B, supports a verified context window of 262,144 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
EXAONE 4.5 33B uses Exaone4_5_ForConditionalGeneration with 33B. The stored architecture profile records 64 layers, hidden width 5120, 40 attention heads and a 262,144 token context ceiling where the official configuration publishes those fields.
Verified Exaone4_5_ForConditionalGeneration architecture
262,144 token certified context
Official BF16, FP8 checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Recommended hardware
Use only the listed certified runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Reference only
Commercial rights are not verified, so this release is not commercially recommendable.
License and self-hosting considerations
The full 33B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
EXAONE 4.5 33B is published under other. The release-scoped official license link and verification date are retained with the catalogue record.
Unknown / Not Recommendable. Commercial rights are not verified, so this release is not commercially recommendable.
Alternatives and sibling variants
EXAONE-4.5-33B-GGUF, EXAONE-4.5-33B-AWQ, EXAONE-4.5-33B-FP8
Release timeline
lg-ai-research/exaone/exaone-4-5-33b
Source details
Source revision: 82680856b53713ab773a163772557f7c01f52c6b21f0060d01ed863d84d6d19c
Frequently asked questions
How much VRAM does EXAONE 4.5 33B require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support EXAONE 4.5 33B?
No runtime is presented as supported until upstream evidence is verified.
Can EXAONE 4.5 33B be used commercially?
The stored commercial classification is Unknown / Not Recommendable. Consult the official other terms for the final obligations.
Official references
Last verified: 2026-09-03