Doubao-pro

Closed weights ByteDance 500B parameters October 2024

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
ByteDance
Organisation type
Industry
Country
China
Published
28 October 2024

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Language modeling/generation, Question answering, Text summarization, Text classification

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
500B

[Speculative] Doubao's large language model has scaled up from 35 billion parameters to 800 billion, with 500 billion and 800 billion parameter models currently under training. https://xueqiu.com/9637001584/309910396?md5__1038=7qmx2DyDuie4cDBqDTQEWqDtMvO4iTphD

Training data
8,350,000,000,000 tokens

[Speculative] Doubao's pre-training data volume is approximately 500TB, with only about 10% of this data actually used for training. The current version employs a non-Mixture-of-Experts (MoE) architecture. In the future, MoE architecture may be introduced to increase parameter count and performance, while also integrating multimodal data solutions. So this model is dense, and the training data is probably all text tokens, not multimodal. 50TB * 167M tokens/GB ~= 8.35 trillion 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
2.5 × 10²⁵ FLOP

6ND = 6 * 500*10^9 * 8350*10^9 = 2.505e+25

How it was established
Operation counting

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Data centre
There is no paper to reference, also no information in media.

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.

Likely above 10²³ FLOP
Yes
Why it is tracked
Training cost
Record confidence
Likely

Sources

Where this record came from and when it was last checked.

Reference
Doubao General Model Pro (Doubao-pro)
Last updated
10 February 2026

What the numbers mean

Where it came from

Doubao-pro was published by ByteDance, in China, in October 2024. It comes out of industry.

It works in Language, and is recorded as doing language modeling/generation, Question answering, Text summarization, Text classification.

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 2.5 × 10²⁵ FLOP of arithmetic, which is a statement about the training budget rather than about inference.

The training set ran to roughly 8,350,000,000,000 tokens.

The reason it appears in this catalogue at all is training cost.

Answers

Doubao-pro — common questions

01

Is Doubao-pro open source?

No. Doubao-pro has not had its weights published, so it exists only as a service controlled by its owner.

02

How many parameters does Doubao-pro have?

Doubao-pro has 500B parameters. [Speculative] Doubao's large language model has scaled up from 35 billion parameters to 800 billion, with 500 billion and 800 billion parameter models currently under training. https://xueqiu.com/9637001584/309910396?md5__1038=7qmx2DyDuie4cDBqDTQEWqDtMvO4iTphD. 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.

03

Who created Doubao-pro?

Doubao-pro was published by ByteDance, based in China, categorised as industry.

04

When was Doubao-pro released?

Doubao-pro 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.

05

What is Doubao-pro used for?

Doubao-pro works in Language, and is recorded as handling language modeling/generation, Question answering, Text summarization, Text classification. These are the areas it was designed around; they describe intent rather than a hard boundary.

06

How much compute was used to train Doubao-pro?

Around 2.5 × 10²⁵ 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.

07

What GPU do I need to run Doubao-pro?

None. Doubao-pro 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.

Source

Original publication

Record last updated 10 February 2026

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.