modelgrep

Google: Gemini 3.6 Flash vs OpenAI: GPT-5.3-Codex

google logoGoogle: Gemini 3.6 Flashgoogle/gemini-3.6-flashVSopenai logoOpenAI: GPT-5.3-Codexopenai/gpt-5.3-codex

Google: Gemini 3.6 Flash wins on more metrics (8 of 9), but the right pick depends on what you optimize for — see the breakdown below.

MetricGoogle: Gemini 3.6 FlashOpenAI: GPT-5.3-Codex
Intelligence Index50.1✓44.3
Coding Index69.2✓—
GPQA Diamond93%✓92%
Design Arena Elo—1198✓
Speed (tokens/sec)126✓—
Latency1.8s✓—
Input price /M$1.50✓$1.75
Output price /M$7.50✓$14.00
Context window1.0M✓400K
CapabilitiesReasoningToolsJSONVisionAudioReasoningToolsJSONVision
Google: Gemini 3.6 Flash details →OpenAI: GPT-5.3-Codex details →