Gemini 3.5 Flash vs gpt-oss-20b
Google vs OpenAI — benchmarks, pricing, and capabilities side by side.
- •Gemini 3.5 Flash has the higher intelligence index (92.2 vs 73.6)
- •gpt-oss-20b is cheaper ($0.03 vs $1.50 per 1M input)
- •gpt-oss-20b is faster
- •Gemini 3.5 Flash has a larger context window (1M)
| Gemini 3.5 Flash | gpt-oss-20b | |
|---|---|---|
| Intelligence index | 92.2 | 73.6 |
| Developer | OpenAI | |
| Type | Multimodal | LLM |
| Access | API only | Open weights |
| Context window | 1,048,576 tokens | 131,072 tokens |
| Input price | $1.50 / 1M | $0.03 / 1M |
| Output price | $9.00 / 1M | $0.14 / 1M |
| Speed | 221 tok/s | 1000 tok/s |
| Released | May 19, 2026 | August 5, 2025 |
| Parameters | — | 21B (3.6B active) |
| Input modalities | Text, Image, Audio, Video | Text |
| Output modalities | Text | Text |
Shared benchmarks
Gemini 3.5 Flash
gpt-oss-20b
GPQA Diamond
92.2
71.5
Humanity’s Last Exam
41
17.3