MegaScale (175B)

Closed weights ByteDance,Peking University 175B parameters February 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,Peking University
Organisation type
Industry,Academia
Country
China
Published
23 February 2024
Authors
Ziheng Jiang, Haibin Lin, Yinmin Zhong, Qi Huang, Yangrui Chen, Zhi Zhang, Yanghua Peng, Xiang Li, Cong Xie, Shibiao Nong, Yulu Jia, Sun He, Hongmin Chen, Zhihao Bai, Qi Hou, Shipeng Yan, Ding Zhou, Yiyao Sheng, Zhuo Jiang, Haohan Xu, Haoran Wei, Zhang Zhang, Pengfei Nie, Leqi Zou, Sida Zhao, Liang Xiang, Zherui Liu, Zhe Li, Xiaoying Jia, Jianxi Ye, Xin Jin, Xin Liu

What it does

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

Domain
Language
Task
Language modeling/generation
Approach
Self-supervised learning

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
175B

Two models are trained for epoch to evaluate the MegaScale training system; one model with 175B and another with 530B parameters. This entry reports the 175B model. There is a third production model mentioned, with fewer details.

Training data
300,000,000,000 tokens

300B tokens * 0.75 words/token = 225B words

Epochs
1

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
2.7 × 10²³ FLOP

Table 2 gives details for the 175B model. Looking at the largest 1 epoch run with 12288 GPUs: 2166.3 aggregate PFlops/sec * 1.75 days * 24 hours/day * 3600 seconds/hour = 3.275e23 This is consistent with the theoretical computation counting estimate, if they factor MFU rate into their 2166.3 figure: 2 × 175B params × 3 × 300B tokens × 1 epoch = 2.29e23 I use the geometric mean of these two: (3.275e23 + 2.29e23) / 2 = 2.74e23

How it was established
Operation counting,Hardware

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
Chips used
12,288
Wall-clock time
42 hours
Hardware utilisation
MFU 55.2%

MFU given in Table 2. Table 3 shows higher utilization when using a smaller number of GPUs.

Power draw
9.7 MW

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

repo, but no training code for the big model https://github.com/volcengine/vescale Model weights 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
SOTA improvement

Improves SOTA in FLOP utilization for distributed LLM training by 1.34X

Record confidence
Confident
Citations
302

Sources

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

Reference
MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs
Last updated
25 May 2026

What the numbers mean

About this model

MegaScale (175B) was published by ByteDance,Peking University, in China, in February 2024. It comes out of industry,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.

Training and provenance

Training it took roughly 2.7 × 10²³ FLOP of computation, on NVIDIA A100 — a measure of what producing the model cost, not of how fast it answers.

Around 300,000,000,000 tokens went into training it.

Its inclusion criterion is sOTA improvement.

Answers

MegaScale (175B) — common questions

01

Is MegaScale (175B) open source?

No. MegaScale (175B) has not had its weights published, so it exists only as a service controlled by its owner.

02

How many parameters does MegaScale (175B) have?

MegaScale (175B) has 175B parameters. Two models are trained for epoch to evaluate the MegaScale training system; one model with 175B and another with 530B parameters. This entry reports the 175B model. There is a third production model mentioned, with fewer details. 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 MegaScale (175B)?

MegaScale (175B) was published by ByteDance,Peking University, based in China, categorised as industry,Academia.

04

When was MegaScale (175B) released?

MegaScale (175B) 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.

05

What is MegaScale (175B) used for?

MegaScale (175B) 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.

06

How much compute was used to train MegaScale (175B)?

Around 2.7 × 10²³ FLOP, on NVIDIA A100. 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 MegaScale (175B)?

None. MegaScale (175B) 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 25 May 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.