TransE
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
- Universite de Technologie de Compiègne – CNRS,Google
- Organisation type
- Academia,Industry
- Country
- France, United States of America
- Published
- 5 December 2013
- Authors
- Antoine Bordes, Nicolas Usunier, Alberto Garcia- Duran, Jason Weston, and Oksana Yakhnenko
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Entity embedding
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
- 942M
- Training data
- 17,500,000 tokens
Based on the TransE architecture, the authors give a formula for how the model size scales with the dimensionality of the dataset. The model scale is proportional to: k*(n_e+n_r) where k is the embeddings dimension, n_e is the number of entities, and n_r is the number of relationships. They studied using the TransE model for two datasets: FB15k and FB1M. The FB15k model has 810000 parameters. FB15k has 14951 entities and 1345 relationships. FB1M has 1000000 entities and 23382 relationships. Th…
"it can be successfully trained on a large scale data set with 1M entities, 25k relationships and more than 17M training samples"
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
- 1.3 × 10¹⁸ FLOP
- How it was established
- Hardware
8 GPUs (they don't specify which, so I used the average for FP32 for 2017 from the write-up table) 8 hours 0.33 util rate
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Compute cost
- $30
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Frontier model
- Yes
- Why it is tracked
- Highly cited
- Record confidence
- Speculative
- Citations
- 8,347
Sources
Where this record came from and when it was last checked.
- Reference
- Translating Embeddings for Modeling Multi- relational Data
- Last updated
- 1 January 2026
What the numbers mean
Where it came from
TransE was published by Universite de Technologie de Compiègne – CNRS,Google, in the country recorded as France, during December 2013. The publishing organisation is categorised as academia,Industry.
It works in the domain of Language, and is recorded as performing the task of entity embedding.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
Training and provenance
Training it took a computation budget of roughly 1.3 × 10¹⁸ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 17,500,000 tokens of text.
Its inclusion criterion: highly cited.
Answers
TransE — common questions
TransE— how many parameters does it have?
It has a parameter count of 942M. Based on the TransE architecture, the authors give a formula for how the model size scales with the dimensionality of the dataset. The model scale is proportional to: k*(n_e+n_r) where k is the embeddings dimension, n_e is the number of entities, and n_r is the number of relationships. They studied using the TransE model for two datasets: FB15k and FB1M. The FB15k model has 810000 parameters. FB15k has 14951 entities and 1345 relationships. FB1M has 1000000 entities and 23382 relationships. Therefore, the FB1M model will be bigger than the FB15k model by a factor of (23382e6)/(14951*1345) => N = 8.1e5 * (23382e6)/(14951*1345) = 942e6. 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.
TransE— who created it?
It was published by Universite de Technologie de Compiègne – CNRS,Google, based in France, an organisation categorised as academia,Industry.
TransE— when was it released?
It was published in December 2013. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
TransE— what is it used for?
It works in the domain of Language, and is recorded as handling the task of entity embedding. 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.
TransE— how much compute was used to train it?
Training consumed around 1.3 × 10¹⁸ FLOP. 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.
TransE— 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.
TransE— is it open source?
The licensing was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
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.