GPT-5.6 Luna vs Gemini 3.8 Flash

Head-to-head technical comparison of intelligence, capabilities, and API pricing.

GPT-5.6 Luna
OpenAI
AI Mastery Index92
LMSYS Arena1455
Technical (MMLU-Pro)92%
Cost per 1M Tokens$0.2
Gemini 3.8 Flash
Google
AI Mastery Index93.5
LMSYS Arena1494
Technical (MMLU-Pro)93.2%
Cost per 1M Tokens$0.75

Which should you use?

Gemini 3.8 Flash leads on the composite index by 1.5 points (92 vs 93.5).

Cost is the sharper difference: Gemini 3.8 Flash is 3.8x the input price of GPT-5.6 Luna ($0.75 against $0.2 per 1M), so at volume the choice is usually decided by budget rather than benchmarks.

Pick GPT-5.6 Luna when

  • Cheaper input tokens — $0.2 against $0.75 per 1M, so Gemini 3.8 Flash costs 3.8x as much to feed.
  • Better value per point of measured capability: $0.002 per index point against $0.008.

Pick Gemini 3.8 Flash when

  • Higher composite index — 93.5 against 92, a gap of 1.5 points.
  • Ahead on Arena Elo by 39 points (1494 vs 1455), which is outside the board's published confidence intervals.
  • Stronger on MMLU-Pro: 93.2% against 92%.

Value per point of measured capability, at list input pricing: GPT-5.6 Luna $0.002 · Gemini 3.8 Flash $0.008 per index point. Run your own token mix through the token cost calculator — a comparison at list price ignores caching and batch discounts, which move real bills more than this gap does.

Our reporting on OpenAI and Google