DeepSeek V4.1 Flash 1017 vs Gemini 3.5 Flash

DeepSeek against Google. Prices, scores and speed below are live.

  • DeepSeek V4.1 Flash 1017 costs 92% less on blended price ($0.39 against $5.06 per 1M tokens, 3 input to 1 output).
  • Gemini 3.5 Flash scores higher on the Artificial Analysis Intelligence Index: 33 against 32.6.
  • On the 21 benchmarks both have scores on, DeepSeek V4.1 Flash 1017 leads on 16 and Gemini 3.5 Flash on 5.
  • Both read up to 1M tokens of context.
  • DeepSeek V4.1 Flash 1017 is newer: listed Oct 17, 2026, against May 19, 2026.

Side by side

DeepSeek V4.1 Flash 1017Gemini 3.5 Flash
LabDeepSeekGoogle
ListedOct 17, 2026May 19, 2026
Input per 1M tokens$0.23$2.25
Output per 1M tokens$0.90$13.50
Cached input per 1M$0.0045$0.23
Blended (3:1) per 1M$0.39$5.06
Context window1M1M
Max output393K66K
Knowledge cutoff-2025-01-01
Intelligence index32.633
Coding index-70.1
ReasoningYesYes
Reads imagesYesYes
Tool useYesYes
Structured outputYesYes
Measured output speed-126 tokens/s

Every benchmark both are scored on

Each model's best score per benchmark, with its source. Scores from different sources or settings are not always comparable; each links to where it was published.

BenchmarkDeepSeek V4.1 Flash 1017Gemini 3.5 Flash
AIME99.4%Epoch AI88.9%Epoch AI
APEX-Agents69.8%Epoch AI27.5%Epoch AI
ARC-AGI-194.5%Epoch AI92.5%Epoch AI
ARC-AGI-278.1%Epoch AI72.1%Epoch AI
Artificial Analysis Intelligence Index32.6Artificial Analysis and OpenRouter33Artificial Analysis and OpenRouter
Chess Puzzles46%Epoch AI45%Epoch AI
Design Arena: 3D1332 EloDesign Arena1213 EloDesign Arena
Design Arena: ASCII Art1222 EloDesign Arena1239 EloDesign Arena
Design Arena: Code1281 EloDesign Arena1250 EloDesign Arena
Design Arena: Data Viz1271 EloDesign Arena1235 EloDesign Arena
Design Arena: Game Dev1248 EloDesign Arena1243 EloDesign Arena
Design Arena: SVG1235 EloDesign Arena1241 EloDesign Arena
Design Arena: UI Component1274 EloDesign Arena1260 EloDesign Arena
Design Arena: Website1289 EloDesign Arena1256 EloDesign Arena
FrontierMath Tier 4 (v2)46.3%Epoch AI26.8%Epoch AI
FrontierMath Tiers 1-3 (v2)71.2%Epoch AI62.8%Epoch AI
GPQA Diamond90.1%Artificial Analysis and OpenRouter92.3%Artificial Analysis and OpenRouter
Mystery Game Puzzles49%Epoch AI28%Epoch AI
SciCode55.9%Epoch AI53.9%Epoch AI
SimpleQA Verified42.5%Epoch AI66.2%Epoch AI
tau-bench (Airline)73.3%Artificial Analysis and OpenRouter72.7%Artificial Analysis and OpenRouter

More comparisons

Where these figures come from

Prices are GenMagic's own, what the studio and the API charge, recomputed from each provider's price list every few minutes. The indices are the Artificial Analysis indices, and every benchmark score links to its source, re-read daily. Measured speed is GenMagic's own generation records, in aggregate. See DeepSeek models, Google models, every LLM's price and our sources.