RoFormer
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
- Zhuiyi Technology
- Organisation type
- Industry
- Country
- China
- Published
- 8 November 2023
- Authors
- Jianlin Su, Yu Lu, Shengfeng Pan, Ahmed Murtadha, Bo Wen, Yunfeng Liu
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Text classification, Language modeling/generation
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
- 110M
- Training data
- 3,276,800,000 tokens
"we use bert-base" BERT base has 110M parameters from https://arxiv.org/pdf/1810.04805
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
- 2.2 × 10¹⁸ FLOP
- How it was established
- Operation counting
6 FLOP/parameter/token * 110000000 parameters * 3276800000 tokens = 2162688000000000000 FLOP
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 V100
- Chips used
- 4
- Power draw
- 2.4 kW
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
- Unreleased
- Training code
- Open source
Apache 2.0 https://github.com/ZhuiyiTechnology/roformer (only Chinese roformer implementations, not BERT) RoFormer is already integrated into Huggingface: https://huggingface.co/docs/transformers/model_doc/roformer
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- Historical significance
- Record confidence
- Confident
the paper introduced RoPE
Sources
Where this record came from and when it was last checked.
- Reference
- RoFormer: Enhanced Transformer with Rotary Position Embedding
- Last updated
- 28 November 2025
What the numbers mean
About this model
RoFormer was published by Zhuiyi Technology, in China, in November 2023. industry is the category the publisher falls under.
It works in Language, and is recorded as doing text classification, 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 roughly 2.2 × 10¹⁸ FLOP of computation, on NVIDIA V100 — a measure of what producing the model cost, not of how fast it answers.
The training set ran to roughly 3,276,800,000 tokens.
Its inclusion criterion is historical significance.
Answers
RoFormer — common questions
How many parameters does RoFormer have?
RoFormer has 110M parameters. "we use bert-base" BERT base has 110M parameters from https://arxiv.org/pdf/1810.04805. 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.
Who created RoFormer?
RoFormer was published by Zhuiyi Technology, based in China, categorised as industry.
When was RoFormer released?
RoFormer was published in November 2023. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is RoFormer used for?
RoFormer works in Language, and is recorded as handling text classification, 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.
How much compute was used to train RoFormer?
Around 2.2 × 10¹⁸ FLOP, on NVIDIA V100. 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.
What GPU do I need to run RoFormer?
None. RoFormer 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.
Is RoFormer open source?
No. RoFormer has not had its weights published, so it exists only as a service controlled by its owner.
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.