Nemotron-4 15B
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
- NVIDIA
- Organisation type
- Industry
- Country
- United States of America
- Published
- 27 February 2024
- Authors
- Jupinder Parmar, Shrimai Prabhumoye, Joseph Jennings, Mostofa Patwary, Sandeep Subramanian, Dan Su, Chen Zhu, Deepak Narayanan, Aastha Jhunjhunwala, Ayush Dattagupta, Vibhu Jawa, Jiwei Liu, Ameya Mahabaleshwarkar, Osvald Nitski, Annika Brundyn, James Maki, Miguel Martinez, Jiaxuan You, John Kamalu, Patrick LeGresley, Denys Fridman, Jared Casper, Ashwath Aithal, Oleksii Kuchaiev, Mohammad Shoeybi, …
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation, Code generation, Question answering, Translation, Quantitative reasoning
- Numerical format
- BF16
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
- 15B
- Training data
- tokens
15b
"15-billion-parameter large multilingual language model trained on 8 trillion text tokens"
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
- 7.5 × 10²³ FLOP
- How it was established
- Operation counting,Hardware
6ND = 6 FLOP/token/parameter * 15*10^9 parameters * 8*10^12 tokens = 7.2e+23 FLOP "Nemotron-4 was trained using 384 DGX H100 nodes; each node contains 8 H100 80GB SXM5 GPUs based on the NVIDIA Hopper architecture (NVIDIA, 2022). Each H100 GPU has a peak throughput of 989 teraFLOP/s when doing 16-bit floating point (bfloat16) arithmetic without sparsity. Table 2 reports more detailed training schedule: 989*10^12 FLOP/sec * 3600 sec/hour * 24 hours * (768 gpus * 0.343 [reported utilization] * 0…
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 H100 SXM5 80GB
- Chips used
- 3,072
- Wall-clock time
- 312 hours (13 days)
- Hardware utilisation
- MFU 30.5%
- Power draw
- 4.3 MW
"Training was completed in approximately 13 calendar days."
MFU shown in Table 2
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
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Likely above 10²³ FLOP
- Yes
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- Nemotron-4 15B Technical Report
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
Nemotron-4 15B was published by NVIDIA, in the country recorded as United States of America, during February 2024. It comes out of an organisation categorised as industry.
It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Code generation, Question answering, Translation, Quantitative reasoning.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Producing it required arithmetic totalling around 7.5 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Answers
Nemotron-4 15B — common questions
Nemotron-4 15B— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Code generation, Question answering, Translation, Quantitative reasoning. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
Nemotron-4 15B— how much compute was used to train it?
Training consumed around 7.5 × 10²³ FLOP, on hardware recorded as NVIDIA H100 SXM5 80GB. 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.
Nemotron-4 15B— 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.
Nemotron-4 15B— is it open source?
No. Its weights have not been published, so it exists only as a service controlled by its owner.
Nemotron-4 15B— how many parameters does it have?
It has a parameter count of 15B. 15b. 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.
Nemotron-4 15B— who created it?
It was published by NVIDIA, based in United States of America, an organisation categorised as industry.
Nemotron-4 15B— when was it released?
It was published in February 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.
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.