In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- GPT-6 Luna is cheaper per output token: $0.50 against $6 per million tokens, so Grok 4.7 costs 12 times as much.
- GPT-6 Luna reads more at once: a context window of 1M tokens against 500K.
- On the 2 independent boards that rank both, Grok 4.7 places higher on 2 and GPT-6 Luna on 0.
Side by side
The specs.
| GPT-6 Luna | Grok 4.7 | |
|---|---|---|
| Made by | OpenAI · US | SpaceXSI (formerly xAI) · US |
| Released | Sep 22, 2026 | Sep 21, 2026 |
| Price per 1M tokens | $0.10 in / $0.50 out | $2 in / $6 out |
| Context window | 1M | 500K |
| Takes in | text, image | text, image |
| Download and run it yourself | No | No |
| Size | Not published | Not published |
| Maker says | “Our most efficient model for focused, high-volume tasks.” | “Grok 4.7 is our most capable model for coding and knowledge work.” |
| Sources | Prices, Oct 20, 2026 · Description | Prices, Oct 20, 2026 · Description |
Independent tests
Where they place.
Position on each board we follow (1 is best). Epoch AI's and Vectara's scores are shown; the other boards are linked for their own figures.
| Board | GPT-6 Luna | Grok 4.7 |
|---|---|---|
| Epoch Capabilities IndexEpoch AI, Oct 20, 2026 | #27156.31 | Not listed |
| LMArena WebDevLMArena, Oct 19, 2026 | Not listed | #18 |
| LMArena Agent ArenaLMArena, Oct 19, 2026 | #19 | #17 |
| Humanity's Last ExamScale AI | Not listed | #13 |
| Vals IndexVals AI, Oct 19, 2026 | #19 | #11 |
| Terminal-Bench 4.0Vals AI, Oct 19, 2026 | Not listed | #20 |
| Vibe Code BenchVals AI, Oct 18, 2026 | Not listed | #12 |
More comparisons