Universal-2-TF
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
- AssemblyAI
- Organisation type
- Industry
- Published
- 30 October 2024
- Authors
- Core Research Luka Chkhetiani (lead), Andrea Vanzo (lead), Yash Khare, Taufiquzzaman Peyash, Ilya Sklyar Research Contributors Michael Liang, Rami Botros, Ruben Bousbib Research Data Ahmed Etefy Benchmarking Pegah Ghahremani, Gabriel Oexle, Jaime Lorenzo Trueba Research Infrastructure William Pipsico Ferreira Production Engineering Ben Gotthold, Soheyl Bahadoori, Mimi Chiang, Aleksandar Mitov…
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Speech
- Task
- Speech recognition (ASR)
- Base model
- Conformer
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
- 600M
- Training data
- tokens
1. ASR decoder: "Our models are based on an ASR decoder architecture called the Recurrent Neural Network Transducer (RNN-T), which offers advantages in scalability, robustness against hallucinations, and timestamp accuracy—key factors for real-world usage at scale. We use a 600M parameter Conformer RNN-T model." 2. Text Formatting module.
1. ASR decoder:: "For Universal-2, we doubled the size of the supervised training dataset from 150,000 hours to 300,000 hours." "we first pre-trained an RNN-T encoder using 12.5 million hours of diverse, multilingual audio. After the pre-training, the encoder was combined with a randomly initialized decoder, and the entire model was fine-tuned using a combination of the supervised dataset described above and a pseudo-labeled dataset, similar to the approach used in Universal-1." 2. Text formatt…
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
- API access
- Training code
- Unreleased
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Unknown
Sources
Where this record came from and when it was last checked.
- Reference
- Robust All-Neural Text Formatting for ASR
- Last updated
- 28 November 2025
What the numbers mean
Background
Universal-2-TF was published by AssemblyAI, in October 2024. It comes out of industry.
It works in Speech, and is recorded as doing speech recognition (ASR).
It is derived from Conformer rather than trained from scratch, which is the usual way a specialised model is produced.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
Answers
Universal-2-TF — common questions
How many parameters does Universal-2-TF have?
Universal-2-TF has 600M parameters. 1. ASR decoder: "Our models are based on an ASR decoder architecture called the Recurrent Neural Network Transducer (RNN-T), which offers advantages in scalability, robustness against hallucinations, and timestamp accuracy—key factors for real-world usage at scale. We use a 600M parameter Conformer RNN-T model." 2. Text Formatting module. 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 Universal-2-TF?
Universal-2-TF was published by AssemblyAI, categorised as industry.
When was Universal-2-TF released?
Universal-2-TF was published in October 2024. 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 Universal-2-TF used for?
Universal-2-TF works in Speech, and is recorded as handling speech recognition (ASR). Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
What GPU do I need to run Universal-2-TF?
None. Universal-2-TF 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 Universal-2-TF open source?
No. Universal-2-TF 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.