Comparing 2 models

DeepSeek V4.1 Flash vs Gemini 3.7 Flash

Benchmarks through

Verdict

DeepSeek V4.1 Flash places higher on 4 of the 4 tasks where both are ranked. DeepSeek V4.1 Flash costs 14.8x less per input token. Only DeepSeek V4.1 Flash has open weights for self-hosting.

  • Choose DeepSeek V4.1 Flash for Value LLM (#1), Fast Summaries (#2), Cheapest Capable LLM (#2), you need to self-host.

Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.

metric
Model 1DeepSeek V4.1 Flash
Model 2Gemini 3.7 Flash
Best for
JSON & Schema Output (#1)
Business Writing (#2)
Provider
deepseek
google
Context
1.0MBest
1.0MBest
Input / 1M
$0.051Best
$0.75
Output / 1M
$0.20Best
$3.75
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
yesBest
—
Released
Sep 10, 2026
Aug 13, 2026

Measured head to head

Benchmarks both models were run on

EvaluationDeepSeek V4.1 FlashGemini 3.7 FlashSource
FrontierMath Tiers 1-3 (Epoch AI)67.4% ± 2.871.6% ± 2.7source · 2026-10-08
GPQA Diamond (Epoch AI)89.8% ± 1.794.8% ± 1.3source · 2026-10-09
LiveBench Agentic Coding77.3%58.3%source · 2026-10-18
LiveBench Coding80.0%78.9%source · 2026-10-18
LiveBench Data Analysis79.3%68.0%source · 2026-10-18
LiveBench Instruction Following70.0%79.9%source · 2026-10-18
LiveBench Language81.2%85.5%source · 2026-10-18
LiveBench Math93.3%93.5%source · 2026-10-18
LiveBench Reasoning86.7%87.8%source · 2026-10-18
LMArena Agent Bash Recovery7.9-5.4source · 2026-10-08
LMArena Creative Writing1438 Elo1485 Elosource · 2026-11-11
LMArena Instruction Following1473 Elo1485 Elosource · 2026-11-11
LMArena Long Query1474 Elo1494 Elosource · 2026-11-11
LMArena Math1479 Elo1506 Elosource · 2026-11-11
LMArena Text1475 Elo1489 Elosource · 2026-11-11
LMArena Text Factuality1468 Elo1476 Elosource · 2026-11-11
LMArena Vision1263 Elo1294 Elosource · 2026-11-09
LMArena Vision Diagrams1296 Elo1320 Elosource · 2026-11-09
LMArena Vision OCR1274 Elo1308 Elosource · 2026-11-09
LMArena WebDev1621 Elo1588 Elosource · 2026-11-11
LMArena WebDev Image-to-Web1610 Elo1600 Elosource · 2026-11-10
OTIS Mock AIME (Epoch AI)98.3% ± 0.697.2% ± 1.9source · 2026-10-08
SimpleQA Verified (Epoch AI)42.9% ± 1.669.2% ± 1.5source · 2026-10-13

Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.