Compare AI models

Specifications
Parameters
671B—
Context window
128k1M
API pricing
Input price
—$0.10
Output price
—$0.40
Cached input price
—$0.01
Cheapest input
$0.256DeepInfra$0.05Google AI Studio
Cheapest output
$0.712DeepInfra$0.20Google AI Studio
$103StreamLake$40.00Google list price
Benchmarks
SWE-Bench Verified
42%31.6%
LiveCodeBench
40.5%33.7%
GPQA Diamond
59.1%64.6%
Overview
CompanyDeepSeekGoogle
Release dateDec 26 2024Jun 17 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek V3 leads Gemini 2.5 Flash-Lite on 2 of the 3 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, GPQA Diamond). Only Gemini 2.5 Flash-Lite has a verified first-party price for an available API: $0.10 per million input tokens and $0.40 per million output tokens. No currently available pay-as-you-go API rate is tracked for DeepSeek V3. DeepSeek V3 shipped 173 days before Gemini 2.5 Flash-Lite, so benchmark comparisons should account for the intervening progress.

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

On SWE-Bench Verified, DeepSeek V3 leads at 42% vs Gemini 2.5 Flash-Lite at 31.6%. On LiveCodeBench, DeepSeek V3 leads at 40.5% vs Gemini 2.5 Flash-Lite at 33.7%. On GPQA Diamond, Gemini 2.5 Flash-Lite leads at 64.6% vs DeepSeek V3 at 59.1%.