Compare AI models

Specifications
Parameters
671B—
Context window
128k1M
API pricing
Input price
—$1.25
Output price
—$10.00
Cached input price
—$0.125
Cheapest input
$0.256DeepInfra$0.625Google AI Studio
Cheapest output
$0.712DeepInfra$5.00Google AI Studio
$103StreamLake$750Google list price
Benchmarks
SWE-Bench Verified
42%63.8%
LiveCodeBench
40.5%70.4%
GPQA Diamond
59.1%84%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024Mar 25 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Gemini 2.5 Pro leads DeepSeek V3 on 3 of the 3 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, GPQA Diamond). Only Gemini 2.5 Pro has a verified first-party price for an available API: $1.25 per million input tokens and $10.00 per million output tokens. No currently available pay-as-you-go API rate is tracked for DeepSeek V3. DeepSeek V3 shipped 89 days before Gemini 2.5 Pro, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V3) vs 1M (Gemini 2.5 Pro). DeepSeek V3 is open weight, while Gemini 2.5 Pro is closed.

On SWE-Bench Verified, Gemini 2.5 Pro leads at 63.8% vs DeepSeek V3 at 42%. On LiveCodeBench, Gemini 2.5 Pro leads at 70.4% vs DeepSeek V3 at 40.5%. On GPQA Diamond, Gemini 2.5 Pro leads at 84% vs DeepSeek V3 at 59.1%.