o4-mini
A lightweight reasoning model released by OpenAI, representing a compact variant of the o-series reasoning models. It is designed to preserve reasoning capabilities while reducing computational cost, making it suitable for applications that require efficient inference without the full resource demands of larger models in the same series.
Timeline (last 90 days)
On the SimpleQA benchmark, o4-mini declines to answer only 1% of questions, with an error rate of 75%.
o4-mini and gpt-5-thinking-mini have almost the same accuracy on SimpleQA, but the latter makes roughly three times fewer errors than the former.
Codex (the 2025 Agent product) is powered under the hood by reasoning models such as o3 and o4-mini
All Facts (3)
On the SimpleQA benchmark, o4-mini declines to answer only 1% of questions, with an error rate of 75%.
50%Unverifiedo4-mini and gpt-5-thinking-mini have almost the same accuracy on SimpleQA, but the latter makes roughly three times fewer errors than the former.
50%UnverifiedCodex (the 2025 Agent product) is powered under the hood by reasoning models such as o3 and o4-mini
50%