NeuroChess
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.
- Published
- 2 December 1994
- Authors
- S. Thrun
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Games
- Task
- Chess
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
- 72.3K
- Training data
- 9,600,000 tokens
"Prior to learning an evaluation function, the model M (175 input, 165 hidden, and 175 output units)" = 58,090 parameters "NeuroChess then learns an evaluation network V (175 input units, 0 to 80 hidden units, and one output units)." = 14,161 parameters Total: 58,090 + 14,161 = 72,251
"is trained using a database of 120,000 expert games."
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
- 8.6 × 10¹¹ FLOP
- How it was established
- Hardware
Lower bound: 0.3*2*24*60*60*1400000=72576000000=7.26e10 Upper bound: 0.3*14*24*60*60*1400000*20=10160640000000=1.02e13 Geometric mean: 858730812676=8.59e11 (speculative) "Thus far, experiments lasted for 2 days to 2 weeks on I to 20 SUN Sparc Stations. " SparcStation has 1.4 MFLOPS (https://ieeexplore.ieee.org/document/63671)
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Frontier model
- Yes
- Why it is tracked
- Historical significance,Highly cited
- Record confidence
- Speculative
Sources
Where this record came from and when it was last checked.
- Reference
- Learning to Play the Game of Chess
- Last updated
- 28 November 2025
What the numbers mean
Background
NeuroChess was published by its authors, in December 1994.
It works in Games, and is recorded as doing chess.
Its weights were never published, so it can only be reached through its provider. No graphics card changes that.
What went into building it
Training it took roughly 8.6 × 10¹¹ FLOP of computation — a measure of what producing the model cost, not of how fast it answers.
It was trained on about 9,600,000 tokens of text.
Its inclusion criterion is historical significance,Highly cited.
Answers
NeuroChess — common questions
When was NeuroChess released?
NeuroChess was published in December 1994. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is NeuroChess used for?
NeuroChess works in Games, and is recorded as handling chess. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
How much compute was used to train NeuroChess?
Around 8.6 × 10¹¹ 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.
What GPU do I need to run NeuroChess?
None. NeuroChess 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.
Is NeuroChess open source?
The licensing for NeuroChess was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does NeuroChess have?
NeuroChess has 72.3K parameters. "Prior to learning an evaluation function, the model M (175 input, 165 hidden, and 175 output units)" = 58,090 parameters "NeuroChess then learns an evaluation network V (175 input units, 0 to 80 hidden units, and one output units)." = 14,161 parameters Total: 58,090 + 14,161 = 72,251. 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.
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.