AI Dictionary › AI Fundamentals

FLOPS

FLOPS stands for Floating Point Operations Per Second and measures the computing speed of a processor or system. Floating point operations are calculations with decimal numbers, the daily bread of artificial intelligence. It is like measuring how many bricks a mason lays per second: more FLOPS, more slices of computation each second. The magnitudes are enormous: teraFLOPS (a trillion) or petaFLOPS (a quadrillion). Beware the ambiguous plural: "FLOPs" without a resolved trailing s instead means the total number of operations needed to train a model, not a speed.

Definition

FLOPS matter because they quantify the power needed to train or run a model. They are the yardstick for estimating costs, times and emissions: training large models takes billions of billions of operations.

Related terms

• Compute
• GPU and TPU
• Parameters
• Deep Learning
• Inference

More in AI Fundamentals

• LLM (Large Language Model)
• Token
• Context Window
• Temperature
• AI Hallucination
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Embedding

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.