In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- Grok 4.7 reads more at once: a context window of 500K tokens against 128K.
- gpt-oss-120b can be downloaded and run on your own machine; Grok 4.7 can't.
- No independent board we follow ranks both models yet.
Side by side
The specs.
| gpt-oss-120b | Grok 4.7 | |
|---|---|---|
| Made by | OpenAI · US | SpaceXSI (formerly xAI) · US |
| Released | Aug 5, 2025 | Sep 21, 2026 |
| Price per 1M tokens | Free to download; hosts set prices | $2 in / $6 out |
| Context window | 128K | 500K |
| Takes in | text | text, image |
| Download and run it yourself | Yes (Apache 2.0) | No |
| Size | 117B parameters, 5.1B active | Not published |
| Maker says | “production, general purpose, high reasoning use cases” | “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-oss-120b | Grok 4.7 |
|---|---|---|
| LMArena WebDevLMArena, Oct 19, 2026 | Not listed | #18 |
| LMArena Agent ArenaLMArena, Oct 19, 2026 | Not listed | #17 |
| Humanity's Last ExamScale AI | Not listed | #13 |
| Vals IndexVals AI, Oct 19, 2026 | Not listed | #11 |
| Terminal-Bench 4.0Vals AI, Oct 19, 2026 | Not listed | #20 |
| Vibe Code BenchVals AI, Oct 18, 2026 | Not listed | #12 |
| Vectara hallucination testVectara, Oct 16, 2026 | #9214.2% | Not listed |
More comparisons