Phi 4 vs GPT-6 Luna
Phi 4 is cheaper on output tokens, while GPT-6 Luna offers a larger context window. Choose Phi 4 or GPT-6 Luna based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Phi 4 | GPT-6 Luna |
|---|---|---|
| Provider | Microsoft | OpenAI |
| Input / 1M tokens | $0.07 | $0.10 |
| Output / 1M tokens | $0.14 | $0.50 |
| Context window | 16K | 1.1M |
| Parameters | — | — |
| Open weights | — | — |
| Released | — | — |
| GPQA Diamond score | 56.1 | 90.5 |
| Epoch Capabilities Index score | 131 | 157 |
| LMArena (Chatbot Arena) Elo score | 1217 | 1389 |
| LMArena WebDev Elo score | — | 1579 |
| LMArena Vision Elo score | — | 1216 |
FAQ
Is Phi 4 or GPT-6 Luna cheaper?
Phi 4 is cheaper on output tokens ($0.14 vs $0.50 per 1M).
Which has the larger context window, Phi 4 or GPT-6 Luna?
GPT-6 Luna has the larger context window (1.1M tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.