Inference

AI fundamentals

Inference is the use of an already-trained model: given an input, it produces its output. Every chatbot answer is an inference. Its cost (latency, energy, price per token) is a central criterion when sizing an AI solution for production.

In practice at Gensai

Gensai optimises inference costs by choosing the right model size, down to the fully local inference of the Oceans installation.

Same theme

Artificial intelligence (AI)Algorithm Data Dataset Model (AI)Machine learning Deep learning Neural network Supervised learning Unsupervised learning Reinforcement learning Classification Regression Prediction Training Parameters Hyperparameters Bias (AI)

Inference

In practice at Gensai

In application

Same theme