Hyena-2 355M
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
- Stanford University,University of Montreal / Université de Montréal,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms)
- Organisation type
- Academia,Academia,Academia
- Country
- United States of America, Canada
- Published
- 21 February 2023
- Authors
- Michael Poli, Stefano Massaroli, Eric Nguyen, Daniel Y. Fu, Tri Dao, Stephen Baccus, Yoshua Bengio, Stefano Ermon, Christopher Ré
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- 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
- 355M
- Training data
- 15,000,000,000 tokens
- Batch size
- 256
Trained for 15B tokens on The Pile
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
- 3.9 × 10¹⁹ FLOP
- How it was established
- Reported,Operation counting
Table 3: 3.93*10^19 FLOP 6 FLOP / parameter / token * 355 * 10^6 parameters * 15 * 10^9 tokens = 3.195e+19 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 A100 SXM4 80 GB
- Chips used
- 8
- Power draw
- 6.4 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
- Unreleased
apache 2.0 https://github.com/HazyResearch/safari (the repo doesn't seem to contain Hyena-2 355M)
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
- Citations
- 497
Sources
Where this record came from and when it was last checked.
- Reference
- Hyena Hierarchy: Towards Larger Convolutional Language Models
- Last updated
- 25 May 2026
What the numbers mean
What this model is
Hyena-2 355M was published by Stanford University,University of Montreal / Université de Montréal,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms), in United States of America, in February 2023. The organisation is categorised as academia,Academia,Academia.
It works in Language, and is recorded as doing language modeling/generation.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
How it was trained
Training it took roughly 3.9 × 10¹⁹ FLOP of computation, on NVIDIA A100 SXM4 80 GB — a measure of what producing the model cost, not of how fast it answers.
The training set ran to roughly 15,000,000,000 tokens.
Answers
Hyena-2 355M — common questions
Who created Hyena-2 355M?
Hyena-2 355M was published by Stanford University,University of Montreal / Université de Montréal,Mila - Quebec AI (originally Montreal Institute for Learning Algorithms), based in United States of America, categorised as academia,Academia,Academia.
When was Hyena-2 355M released?
Hyena-2 355M was published in February 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 Hyena-2 355M used for?
Hyena-2 355M works in Language, and is recorded as handling language modeling/generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
How much compute was used to train Hyena-2 355M?
Around 3.9 × 10¹⁹ FLOP, on NVIDIA A100 SXM4 80 GB. 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 Hyena-2 355M?
None. Hyena-2 355M 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 Hyena-2 355M open source?
No. Hyena-2 355M has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does Hyena-2 355M have?
Hyena-2 355M has 355M parameters. 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.