I’ve been trying to get GPT-4o via the Chat Completion API to give the same response given the same API arguments without success. Setting a seed and temperature to 0 has resulted in slightly more deterministic behavior but large amounts of randomness still exist. Experiment: 10 API calls to GPT-4o…

Deterministic Results Impossible for GPT-4o

zach-humata December 18, 2024, 5:32pm 5

Appreciate the suggestion but I’m looking for ways to make LLMs deterministic in their response, not a hash and cache approach

Topic		Replies	Views
Seeking Assistance on Achieving Determinism in OpenAI Models Community gpt-4 , api	2	921	August 6, 2024
ChatCompletions are not deterministic even with seed set, temperature=0, top_p=0, n=1 API gpt-4 , api	9	1973	October 7, 2024
Achieving deterministic API output on language models - HOWTO API statistics	3	9584	December 21, 2023
Lack of determinisim even with temp 0 and fixed seed Bugs gpt-4	1	1303	May 24, 2024
Finetuned gpt4o is not deterministic API	5	358	August 26, 2025