Google: Gemini 3.5 Flash-Lite vs OpenAI: GPT-5 Codex
Google: Gemini 3.5 Flash-Lite wins on more metrics (8 of 8), but the right pick depends on what you optimize for — see the breakdown below.
| Metric | Google: Gemini 3.5 Flash-Lite | OpenAI: GPT-5 Codex |
|---|---|---|
| Intelligence Index | 36.5✓ | 36.1 |
| Coding Index | 49.3✓ | — |
| GPQA Diamond | 84%✓ | 84% |
| Design Arena Elo | — | — |
| Speed (tokens/sec) | 341✓ | — |
| Latency | 880ms✓ | — |
| Input price /M | $0.300✓ | $1.25 |
| Output price /M | $2.50✓ | $10.00 |
| Context window | 1.0M✓ | 400K |
| Capabilities | ReasoningToolsJSONVisionAudio | ReasoningToolsJSONVision |