Gemini 2.5 Flash Preview (thinking) vs OpenAI o3
Compare Gemini 2.5 Flash Preview (thinking) by Google AI against OpenAI o3 by OpenAI, tested across 20 shared challenges. Updated May 2026.
Which is better, Gemini 2.5 Flash Preview (thinking) or OpenAI o3?
Gemini 2.5 Flash Preview (thinking) and OpenAI o3 are both competitive models. Gemini 2.5 Flash Preview (thinking) costs $0.175/M input tokens vs $10/M for OpenAI o3. Compare their real outputs side by side below.
Key Differences Between Gemini 2.5 Flash Preview (thinking) and OpenAI o3
Gemini 2.5 Flash Preview (thinking) is made by google while OpenAI o3 is from openai. On pricing, Gemini 2.5 Flash Preview (thinking) costs $0.175/M input tokens vs $10/M for OpenAI o3.
Per 1M tokens
Gemini 2.5 Flash Preview (thinking) is cheaper on both — 57× input, 11× output
No community votes yet. On paper, OpenAI o3 has the edge — bigger model tier.
Gemini 2.5 Flash Preview (thinking) is 11x cheaper per token — worth considering if cost matters.
Style Comparison
OpenAI o3 uses 37.9x more headings
