In short
How they differ.
Worked out from the figures below. Prices and specs are the makers' own, read Oct 20, 2026.
- Gemini 3.8 Flash reads more at once: a context window of 1M tokens against 128K.
- gpt-oss-120b can be downloaded and run on your own machine; Gemini 3.8 Flash can't.
- No independent board we follow ranks both models yet.
Side by side
The specs.
| Gemini 3.8 Flash | gpt-oss-120b | |
|---|---|---|
| Made by | Google · US | OpenAI · US |
| Released | Sep 2, 2026 | Aug 5, 2025 |
| Price per 1M tokens | $0.75 in / $3.75 out (introductory price) | Free to download; hosts set prices |
| Context window | 1M | 128K |
| Takes in | text, image, video, audio, pdf | text |
| Download and run it yourself | No | Yes (Apache 2.0) |
| Size | Not published | 117B parameters, 5.1B active |
| Maker says | “Gemini 3.8 Flash is our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows—all with the speed and cost efficiency of Flash.” | “production, general purpose, high reasoning use cases” |
| 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 | Gemini 3.8 Flash | gpt-oss-120b |
|---|---|---|
| Epoch Capabilities IndexEpoch AI, Oct 20, 2026 | #19157.2 | Not listed |
| LMArena textLMArena, Oct 15, 2026 | #9 | Not listed |
| LMArena creative writingLMArena, Oct 15, 2026 | #7 | Not listed |
| LMArena mathLMArena, Oct 15, 2026 | #4 | Not listed |
| LMArena visionLMArena, Oct 19, 2026 | #12 | Not listed |
| LMArena Agent ArenaLMArena, Oct 19, 2026 | #15 | Not listed |
| Humanity's Last ExamScale AI | #8 | Not listed |
| SWE-Bench ProScale AI, Sep 22, 2026 | #12 | Not listed |
| Remote Labor IndexScale AI | #7 | Not listed |
| Vals IndexVals AI, Oct 19, 2026 | #13 | Not listed |
| GPQA Diamond (run by Vals AI, archived)Vals AI, Sep 1, 2026 | #4 | Not listed |
| MMMU Pro (run by Vals AI, archived)Vals AI, Sep 1, 2026 | #4 | Not listed |
| CUA-bench (games)Vals AI, Oct 13, 2026 | #9 | Not listed |
| Vectara hallucination testVectara, Oct 16, 2026 | Not listed | #9214.2% |
More comparisons