AFM-server

Closed weights Apple July 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
Apple
Organisation type
Industry
Country
United States of America
Published
29 July 2024
Authors
Andy Narayanan, Aonan Zhang, Bowen Zhang, Chen Chen, Chong Wang, Chung-Cheng Chiu, David Qiu, Deepak Gopinath, Dian Ang Yap, Dong Yin, Feng Nan, Floris Weers, Guoli Yin, Haoshuo Huang, Jianyu Wang, Jiarui Lu, John Peebles, Ke Ye, Mark Lee, Nan Du, Qibin Chen, Quentin Keunebroek, Ruoming Pang, Sam Wiseman, Syd Evans, Tao Lei, Tom Gunter, Vivek Rathod, Xiang Kong, Xianzhi Du, Yanghao Li, Yongqiang W…

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
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.

Training data
7,400,000,000,000 tokens

Not explicitly mentioned, but I assume the 7.4T tokens do not involve multiple epochs.

Epochs
1
Batch size
18,949,753

Main pretraining uses sequence length of 4096 tokens; 4096 sequences per batch. During the "continued" pre-training stage, sequence length is upped to 8192 while batch size remains 4096. During context lengthening, sequence length is upped to 32768 while "the recipe is similar to that used for continued pre-training" implies same batch size of 4096. Weighting batch sizes by number of tokens seen in each stage: exp((6.3T * ln(4096 * 4096) + 1T * ln(8192 * 4096) + 100B * ln(32768 * 4096))/ 7.4T)…

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
4.3 × 10²⁴ FLOP

"The AFM base models are dense decoder-only models that build on the Transformer architecture" "We train AFM-server from scratch for 6.3T tokens on 8192 TPUv4 chips, using a sequence length of 4096 and a batch-size of 4096 sequences." "For both models we perform continued pre-training at a sequence length of 8192, with another 1T tokens from a mixture that upweights math and code, and down-weights the bulk web-crawl." "The sustained model-flop-utilization (MFU) for this training run was appro…

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
Google TPU v4
Chips used
8,192
Hardware utilisation
MFU 52.0%

"AFM-server was trained on 8192 TPUv4 chips provisioned as 8 × 1024 chip slices, where slices are connected together by the data-center network (DCN) [Chowdhery et al., 2022]. Only data-parallelism crosses the slice boundary, other types of state sharding are within-slice only as the within-slice interconnect bandwidth is orders of magnitude higher than the DCN. The sustained model-flop-utilization (MFU) for this training run was approximately 52%."

Power draw
5.5 MW
Cloud vendor
Google Cloud
Data centre
Google us-central2-b

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
Hosted access (no API)
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
Significant use

Currently in beta access only, but will be integrated into millions or billions of iPhones.

Record confidence
Likely

Sources

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

Reference
Apple Intelligence Foundation Language Models
Last updated
28 November 2025

What the numbers mean

Where it came from

AFM-server was published by Apple, in United States of America, in July 2024. The organisation is categorised as industry.

It works in Language, and is recorded as doing language modeling/generation.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

Training and provenance

Producing it required around 4.3 × 10²⁴ FLOP of arithmetic, on Google TPU v4, which is a statement about the training budget rather than about inference.

Around 7,400,000,000,000 tokens went into training it.

Its inclusion criterion is significant use.

Answers

AFM-server — common questions

01

Who created AFM-server?

AFM-server was published by Apple, based in United States of America, categorised as industry.

02

When was AFM-server released?

AFM-server was published in July 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.

03

What is AFM-server used for?

AFM-server works in Language, and is recorded as handling language modeling/generation. These are the areas it was designed around; they describe intent rather than a hard boundary.

04

How much compute was used to train AFM-server?

Around 4.3 × 10²⁴ 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.

05

What GPU do I need to run AFM-server?

None. AFM-server 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.

06

Is AFM-server open source?

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

07

How many parameters does AFM-server have?

No parameter count has been published for AFM-server, which is why no memory or speed figure appears on this page.

Source

Original publication

Record last updated 28 November 2025

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.