ModelRefs / Random Seed — AI Glossary
Random Seed — AI Glossary
An integer seed controlling the random number generator for sampling, enabling reproducible model outputs. Also called inference seed.
Overview
Setting a fixed seed makes sampling deterministic for the same model, prompt, and parameters. OpenAI and compatible APIs expose `seed`; the response header includes `system_fingerprint` to detect infrastructure changes that break reproducibility. Essential for regression testing, benchmark reproducibility, and auditing.
Reference details
| Topic | inference |
|---|---|
| Also known as | inference seed |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Random Seed — AI Glossary.
Frequently asked questions
What is Random Seed?
An integer seed controlling the random number generator for sampling, enabling reproducible model outputs.
Is Random Seed the same as inference seed?
Yes — inference seed are common aliases for Random Seed.
What concepts are related to Random Seed?
Closely related concepts include sampling, temperature, logprobs.