DeepSeek V4 Pro vs Gemini 2.5 Pro
DeepSeek V4 Pro edges ahead on overall intelligence (RunFree 72.4 vs 22.3). Here's how they stack up on benchmarks, price and specs.
DeepSeek V4 Pro
DeepSeek
Gemini 2.5 Pro
Google
Shared benchmarks (4)
Loading chart…
| DeepSeek V4 Pro | Gemini 2.5 Pro | |
|---|---|---|
| RunFree Score | 72.4 | 22.3 |
| Blended price / 1M | $0.54 | $3.44 |
| Input / 1M | $0.43 | $1.25 |
| Output / 1M | $0.87 | $10.00 |
| Context window | 1.0M | 1.0M |
| Max output | 384K | 66K |
| Reasoning model | Yes | Yes |
| GPQA Diamond | 90.1% | 86.4% |
| Humanity's Last Exam | 37.7% | 21.6% |
| LiveCodeBench | 93.5% | 69% |
| SWE-Bench Verified | 80.6% | 59.6% |
| Wins | 9 | 0 |
A blank means the metric is tied or one model has no verified data. Prices are blended 3:1 input:output.
Compare more: DeepSeek V4 Pro card · Gemini 2.5 Pro card · full leaderboard