Discriminator Guidance

Open weights Korea Advanced Institute of Science and Technology (KAIST),NAVER November 2022

No estimate

No hardware requirements for this model

This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.

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
Korea Advanced Institute of Science and Technology (KAIST),NAVER
Organisation type
Academia,Industry
Country
Korea (Republic of)
Published
28 November 2022
Authors
Dongjun Kim, Yeongmin Kim, Se Jung Kwon, Wanmo Kang, Il-Chul Moon

What it does

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

Domain
Image generation
Task
Image generation
Numerical format
FP16

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
327,978,752 tokens
Epochs
10

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
2.2 × 10²⁰ FLOP

481 hours * 312 TFLOPS (A100) * 40% utilization

How it was established
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 PCIe
Wall-clock time
481 hours (20 days)

Table 6

Compute cost
$338

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
Open — downloadable
Model access
Open weights (non-commercial)
Training code
Open (non-commercial)

https://github.com/alsdudrla10/DG Attribution-NonCommercial-ShareAlike 4.0 International checkpoints and train code here: https://github.com/alsdudrla10/DG/blob/main/README.md

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
SOTA improvement

"Using our algorithm, we achive state-of-the-art results on ImageNet 256x256 with FID 1.83 and recall 0.64, similar to the validation data's FID (1.68) and recall (0.66)." https://paperswithcode.com/paper/refining-generative-process-with

Record confidence
Confident
Citations
110

Sources

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

Reference
Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
Last updated
25 May 2026

What the numbers mean

About this model

Discriminator Guidance was published by Korea Advanced Institute of Science and Technology (KAIST),NAVER, in Korea (Republic of), in November 2022. The organisation is categorised as academia,Industry.

It works in Image generation, and is recorded as doing image generation.

The weights being open is what puts this page in the calculator rather than only in the catalogue: it is a model you can actually hold.

Training and provenance

The training run consumed about 2.2 × 10²⁰ FLOP, on NVIDIA A100 PCIe. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

Around 327,978,752 tokens went into training it.

Its inclusion criterion is sOTA improvement.

Answers

Discriminator Guidance — common questions

01

What is Discriminator Guidance used for?

Discriminator Guidance works in Image generation, and is recorded as handling image generation. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

02

Where can I download Discriminator Guidance?

The weights for Discriminator Guidance are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.

03

How much compute was used to train Discriminator Guidance?

Around 2.2 × 10²⁰ FLOP, on NVIDIA A100 PCIe. 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.

04

What GPU do I need to run Discriminator Guidance?

We cannot say. Discriminator Guidance has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.

05

Is Discriminator Guidance open source?

Its weights are published, so Discriminator Guidance can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.

06

How many parameters does Discriminator Guidance have?

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

07

Who created Discriminator Guidance?

Discriminator Guidance was published by Korea Advanced Institute of Science and Technology (KAIST),NAVER, based in Korea (Republic of), categorised as academia,Industry.

08

When was Discriminator Guidance released?

Discriminator Guidance was published in November 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

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.