DeepSeek-V4.1-FlashvsGemini 3.8 Flash

DeepSeek-V4.1-Flash
Gemini 3.8 Flash
Specifications
Context window
1M
API pricing
Cheapest input
$0.375Google AI Studio
Cheapest output
$1.875Google AI Studio
Benchmarks
DeepSWE 1.1
74.2%71%
Terminal-Bench 4.0
31.2%19.1%
Terminal-Bench 2.1
90.6%89.4%
Benchmarks
NL2Repo-Bench
65.4%
Terminal-Bench 3.0
30%
CyberGym
88.1%
Humanity's Last Exam · no tools
36.8%
Humanity's Last Exam · with tools
63.9%
Humanity's Last Exam (Verified)
54.9%
BioMysteryBench · hard
56.5%
BioMysteryBench · human solved
88.8%
LAB-Bench 2
86.2%
GPQA Diamond
90.9%
OSWorld 2.0
59%
Agent's Last Exam · pass@1
31.8%
AutomationBench
54.8%
Finance Agent v2
61.4%
Harvey's Legal Agent Benchmark
10%
GDPval-AA v2
1545
CharXiv Reasoning
86.2%
Chartography · with tools
78.9%
GDP.PDF
35%
LVBench
87.1%
LVBench · agentic
87.8%
threejseval
1693
Overview
CompanyDeepSeekGoogle
Release dateSep 10 2026Sep 2 2026
AccessProprietaryProprietary

Other comparisons

DeepSeek-V4.1-FlashvsClaude Fable 5.1Gemini 3.8 FlashvsClaude Fable 5.1DeepSeek-V4.1-FlashvsGPT-6 AstraGemini 3.8 FlashvsGPT-6 AstraDeepSeek-V4.1-FlashvsMuse Spark 1.3Gemini 3.8 FlashvsMuse Spark 1.3DeepSeek-V4.1-FlashvsGrok 4.6Gemini 3.8 FlashvsGrok 4.6DeepSeek-V4.1-FlashvsMistral Medium 3.5Gemini 3.8 FlashvsMistral Medium 3.5DeepSeek-V4.1-FlashvsKimi K3Gemini 3.8 FlashvsKimi K3

Frequently asked questions

DeepSeek-V4.1-Flash leads Gemini 3.8 Flash on 3 of the 3 benchmarks they both report (DeepSWE 1.1, Terminal-Bench 4.0, Terminal-Bench 2.1). Gemini 3.8 Flash shipped 8 days before DeepSeek-V4.1-Flash, so benchmark comparisons should account for the intervening progress.

Published specifications for these two models are limited — see each model page for the latest details.

On DeepSWE 1.1, DeepSeek-V4.1-Flash leads at 74.2% vs Gemini 3.8 Flash at 71%. On Terminal-Bench 4.0, DeepSeek-V4.1-Flash leads at 31.2% vs Gemini 3.8 Flash at 19.1%. On Terminal-Bench 2.1, DeepSeek-V4.1-Flash leads at 90.6% vs Gemini 3.8 Flash at 89.4%.