System 11
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
- Carnegie Mellon University (CMU)
- Organisation type
- Academia
- Country
- United States of America
- Published
- 18 June 1996
- Authors
- HA Rowley, S Baluja, T Kanade
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Vision
- Task
- Face detection
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
- 6.5K
- Training data
- 23,750 tokens
System 11 is a combination of Network 1 and Network 2 Network 1 has 2095 connections and network 2 has 4357 connections (see table 1)
"A typical training run selects approximately 8000 non-face images from the 146,212,178 subimages that are available at all locations and scales in the training scenery images." "Nearly 1050 face examples were gathered from face databases at CMU and Harvard [...] In the training set,15 face examples are generated from each original image [...]" "Create an initial set of non-face images by generating 1000 images with random pixel intensities"
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
- 2.6 × 10¹⁰ FLOP
- How it was established
- Operation counting
Since there is no parameter sharing, the forward compute is roughly twice that of the number of parameters. We use a 2:1 forward-backward ratio as this is a shallow network, with most connections in the first layer. Number of passes (Section 2.1): * "Nearly 1,050 face examples were gathered from face databases [...]" * "Fifteen face examples are generated for the training set from each original image" Training loop: 1. "initial set of nonface images by generating 1,000 random images" 2. Train …
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
- Highly cited
- Record confidence
- Confident
- Citations
- 6,011
Sources
Where this record came from and when it was last checked.
- Reference
- Neural Network-Based Face Detection
- Last updated
- 28 November 2025
What the numbers mean
Where it came from
System 11 was published by Carnegie Mellon University (CMU), in the country recorded as United States of America, during June 1996. The category the publisher falls under is academia.
It works in the domain of Vision, and is recorded as performing the task of face detection.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
Producing it required arithmetic totalling around 2.6 × 10¹⁰ FLOP. That figure measures what producing the model cost, and has no bearing on how fast it answers.
Training consumed a corpus of around 23,750 tokens of text.
The reason it appears in this catalogue at all: highly cited.
Answers
System 11 — common questions
System 11— how many parameters does it have?
It has a parameter count of 6.5K. System 11 is a combination of Network 1 and Network 2 Network 1 has 2095 connections and network 2 has 4357 connections (see table 1). 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.
System 11— who created it?
It was published by Carnegie Mellon University (CMU), based in United States of America, an organisation categorised as academia.
System 11— when was it released?
It was published in June 1996. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
System 11— what is it used for?
It works in the domain of Vision, and is recorded as handling the task of face detection. These are the areas it was designed around; they describe intent rather than a hard boundary.
System 11— how much compute was used to train it?
Training consumed around 2.6 × 10¹⁰ 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.
System 11— what GPU do I need to run it?
None. This 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.
System 11— is it open source?
The licensing was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
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.