HyperCLOVA 204B
No estimate
No hardware requirements for this model
The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- NAVER
- Organisation type
- Industry
- Country
- Korea (Republic of)
- Published
- 10 September 2021
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation
- Approach
- Self-supervised learning
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 204B
- Training data
- 560,000,000,000 tokens
https://www.navercorp.com/navercorp_/ir/announce/2023/NAVER_CEO%20letter%20to%20shareholders_Aug%202023_Eng.pdf
https://twitter.com/arankomatsuzaki/status/1397583304610783238 https://venturebeat.com/ai/naver-trained-a-gpt-3-like-korean-language-model/
Training compute
The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.
- Training compute
- 2 × 10²³ FLOP
Estimations for 82B model (marked as lower bound estimations) "For experiments in Section 4, the model trained with 150B is used for fair comparison, because not all models are finished training at the same iteration. However, experiments in Section 5.2 use the model trained with 300B tokens, as HyperCLOVA Studio provided the 39B and 82B models trained with 300B tokens." 82e9 connections * 2 FLOP/connection * 300e9 tokens * 3 backward pass = 1.476e23 FLOP Calculation using GPU time corroborat…
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Training hardware
- NVIDIA A100
- Compute cost
- $441,770
Availability
Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.
- Weights
- Closed — provider access only
- Model access
- Hosted access (no API)
- Training code
- Unreleased
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Frontier model
- Yes
- Foundation model
- Yes
- Likely above 10²³ FLOP
- Yes
- Why it is tracked
- SOTA improvement
- Record confidence
- Speculative
- Citations
- 92
"HyperCLOVA with our training configuration shows state-of-the-art in-context zero-shot and few-shot learning performances on various downstream tasks in Korean" no evaluations on any standard benchmarks are reported
Sources
Where this record came from and when it was last checked.
- Reference
- South Korea's Naver unveils 'hyperscale AI' platform, language model with more parameters than GPT-3
- Last updated
- 28 November 2025
What the numbers mean
About this model
HyperCLOVA 204B was published by NAVER, in the country recorded as Korea (Republic of), during September 2021. The category the publisher falls under is industry.
It works in the domain of Language, and is recorded as performing the task of language modeling/generation.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
What went into building it
Training it took a computation budget of roughly 2 × 10²³ FLOP, on hardware recorded as NVIDIA A100. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 560,000,000,000 tokens of text.
It is tracked in the underlying dataset for one reason in particular: sOTA improvement.
Answers
HyperCLOVA 204B — common questions
HyperCLOVA 204B— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
HyperCLOVA 204B— how many parameters does it have?
It has a parameter count of 204B. https://www.navercorp.com/navercorp_/ir/announce/2023/NAVER_CEO%20letter%20to%20shareholders_Aug%202023_Eng.pdf. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
HyperCLOVA 204B— who created it?
It was published by NAVER, based in Korea (Republic of), an organisation categorised as industry.
HyperCLOVA 204B— when was it released?
It was published in September 2021. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
HyperCLOVA 204B— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling/generation. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.
HyperCLOVA 204B— how much compute was used to train it?
Training consumed around 2 × 10²³ FLOP, on hardware recorded as NVIDIA A100. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.
HyperCLOVA 204B— what GPU do I need to run it?
None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.