TrellisNet
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
- Carnegie Mellon University (CMU),Bosch Center for Artificial Intelligence,Intel Labs
- Organisation type
- Academia,Industry,Industry
- Country
- United States of America, Germany
- Published
- 15 October 2018
- Authors
- Shaojie Bai, J. Zico Kolter, Vladlen Koltun
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling
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
- 180M
- Training data
- 103,000,000 tokens
- Epochs
- 25
180M, Table 2
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
- 2.8 × 10¹⁸ FLOP
- How it was established
- Operation counting
6 FLOP / parameter / token * 180000000 parameters * 103000000 tokens * 25 epochs = 2.781e+18 FLOP
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
MIT license for code: https://github.com/locuslab/trellisnet/tree/master/TrellisNet/word_WT103
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
- SOTA improvement
- Record confidence
- Confident
- Citations
- 164
- Benchmark data
- TrellisNet
"Experiments demonstrate that trellis networks outperform the current state of the art methods on a variety of challenging benchmarks, including word-level language modeling and character-level language modeling tasks" "On word-level WikiText-103, a trellis network outperforms by 7.6% in perplexity the contemporaneous self-attention-based Relational Memory Core (Santoro et al., 2018), and by 11.5% the work of Merity et al. (2018a). (Concurrently with our work, Dai et al. (2019) employ a transfo…
Sources
Where this record came from and when it was last checked.
- Reference
- Trellis Networks for Sequence Modeling
- Last updated
- 25 May 2026
What the numbers mean
What this model is
TrellisNet was published by Carnegie Mellon University (CMU),Bosch Center for Artificial Intelligence,Intel Labs, in United States of America, in October 2018. The organisation is categorised as academia,Industry,Industry.
It works in Language, and is recorded as doing language modeling.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
How it was trained
Producing it required around 2.8 × 10¹⁸ FLOP of arithmetic, which is a statement about the training budget rather than about inference.
It was trained on about 103,000,000 tokens of text.
The reason it appears in this catalogue at all is sOTA improvement.
Answers
TrellisNet — common questions
Who created TrellisNet?
TrellisNet was published by Carnegie Mellon University (CMU),Bosch Center for Artificial Intelligence,Intel Labs, based in United States of America, categorised as academia,Industry,Industry.
When was TrellisNet released?
TrellisNet was published in October 2018. 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 TrellisNet used for?
TrellisNet works in Language, and is recorded as handling language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
How much compute was used to train TrellisNet?
Around 2.8 × 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.
What GPU do I need to run TrellisNet?
None. TrellisNet 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 TrellisNet open source?
No. TrellisNet has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does TrellisNet have?
TrellisNet has 180M parameters. 180M, Table 2. 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.
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.