Gemini 3 Flash vs Phi-4-multimodal-instruct
Google vs Microsoft — benchmarks, pricing, and capabilities side by side.
- •Gemini 3 Flash has the higher intelligence index (90.2 vs 46.2)
- •Phi-4-multimodal-instruct is cheaper ($0.05 vs $0.50 per 1M input)
- •Gemini 3 Flash is faster
- •Gemini 3 Flash has a larger context window (1M)
| Gemini 3 Flash | Phi-4-multimodal-instruct | |
|---|---|---|
| Intelligence index | 90.2 | 46.2 |
| Developer | Microsoft | |
| Type | Multimodal | Multimodal |
| Access | API only | Open weights |
| Context window | 1,048,576 tokens | 128,000 tokens |
| Input price | $0.50 / 1M | $0.05 / 1M |
| Output price | $3.00 / 1M | $0.10 / 1M |
| Speed | 191 tok/s | 25 tok/s |
| Released | December 17, 2025 | February 1, 2025 |
| Parameters | — | 5600000000 |
| Input modalities | Text, Image, Audio, Video | — |
| Output modalities | Text | — |
Shared benchmarks
Gemini 3 Flash
Phi-4-multimodal-instruct
GPQA Diamond
90.4
31.5
Humanity’s Last Exam
34.7
4.4
LiveCodeBench
90.8
13.1
MMLU-Pro
89
48.5