Compare AI models

Specifications
Parameters
671B—
Context window
128k1M
API pricing
Cheapest input
$0.2088GMICloud—
Cheapest output
$0.3096GMICloud—
$57.24GMICloud · fp8—
Benchmarks
BullshitBench v2
13%20%
Toolathlon
35.2%56.5%
Overview
CompanyDeepSeekGoogle
Release dateDec 1 2025May 19 2026
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Gemini 3.5 Flash leads DeepSeek-V3.2 on 2 of the 2 benchmarks they both report (BullshitBench v2, Toolathlon). DeepSeek-V3.2 shipped 169 days before Gemini 3.5 Flash, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek-V3.2) vs 1M (Gemini 3.5 Flash). DeepSeek-V3.2 is open weight, while Gemini 3.5 Flash is closed.

On BullshitBench v2, Gemini 3.5 Flash leads at 20% vs DeepSeek-V3.2 at 13%. On Toolathlon, Gemini 3.5 Flash leads at 56.5% vs DeepSeek-V3.2 at 35.2%.