DeepSeek-V4-Pro vs Gemini 3.5 Flash
DeepSeek vs Google — benchmarks, pricing, and capabilities side by side.
- •Gemini 3.5 Flash has the higher intelligence index (92.2 vs 88.2)
- •DeepSeek-V4-Pro is cheaper ($0.44 vs $1.50 per 1M input)
- •Gemini 3.5 Flash is faster
| DeepSeek-V4-Pro | Gemini 3.5 Flash | |
|---|---|---|
| Intelligence index | 88.2 | 92.2 |
| Developer | DeepSeek | |
| Type | LLM | Multimodal |
| Access | Open weights | API only |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Input price | $0.44 / 1M | $1.50 / 1M |
| Output price | $0.87 / 1M | $9.00 / 1M |
| Speed | 30 tok/s | 221 tok/s |
| Released | April 24, 2026 | May 19, 2026 |
| Parameters | 1.6T (49B active) | — |
| Input modalities | Text | Text, Image, Audio, Video |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek-V4-Pro
Gemini 3.5 Flash
GPQA Diamond
90.1
92.2
Humanity’s Last Exam
35.9
41