Gemini 1.0 Ultra
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
- Google DeepMind
- Organisation type
- Industry
- Country
- United States of America
- Published
- 6 December 2023
- Authors
- Gemini Team
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Multimodal, Language, Vision
- Task
- Language modeling, Visual question answering, Chat, Translation
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.
- Training data
- 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
- 5 × 10²⁵ FLOP
- How it was established
- Benchmarks,Hardware
This number is an estimate based on limited evidence. In particular, we combine information about the performance of Gemini Ultra on various benchmarks compared to other models, and guesstimates about the hardware setup used for training to arrive at our estimate. Our reasoning and calculations are detailed in this Colab notebook. https://colab.research.google.com/drive/1sfG91UfiYpEYnj_xB5YRy07T5dv-9O_c
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
- Google TPU v4
- Chips used
- 57,000
- Chip-hours
- 132,000,000
- Wall-clock time
- 2,400 hours (100 days)
- Power draw
- 38.4 MW
- Compute cost
- $30,719,420
Dylan Patel, author of SemiAnalysis, speculates that the training duration of Gemini may have been 100 days.
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
API access: https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Frontier model
- Yes
- Likely above 10²³ FLOP
- Yes
- Why it is tracked
- SOTA improvement,Training cost
- Record confidence
- Speculative
- Citations
- 633
"Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks — notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined." Table 2
Sources
Where this record came from and when it was last checked.
- Reference
- Gemini: A Family of Highly Capable Multimodal Models
- Last updated
- 16 December 2025
What the numbers mean
Where it came from
Gemini 1.0 Ultra was published by Google DeepMind, in United States of America, in December 2023. industry is the category the publisher falls under.
It works in Multimodal, Language, Vision, and is recorded as doing language modeling, Visual question answering, Chat, Translation.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
What went into building it
Training it took roughly 5 × 10²⁵ FLOP of computation, on Google TPU v4 — a measure of what producing the model cost, not of how fast it answers.
The reason it appears in this catalogue at all is sOTA improvement,Training cost.
Answers
Gemini 1.0 Ultra — common questions
What GPU do I need to run Gemini 1.0 Ultra?
None. Gemini 1.0 Ultra 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 Gemini 1.0 Ultra open source?
No. Gemini 1.0 Ultra has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does Gemini 1.0 Ultra have?
No parameter count has been published for Gemini 1.0 Ultra, which is why no memory or speed figure appears on this page.
Who created Gemini 1.0 Ultra?
Gemini 1.0 Ultra was published by Google DeepMind, based in United States of America, categorised as industry.
When was Gemini 1.0 Ultra released?
Gemini 1.0 Ultra was published in December 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 Gemini 1.0 Ultra used for?
Gemini 1.0 Ultra works in Multimodal, Language, Vision, and is recorded as handling language modeling, Visual question answering, Chat, Translation. 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.
How much compute was used to train Gemini 1.0 Ultra?
Around 5 × 10²⁵ FLOP, on Google TPU v4. 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.
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.