Comparing 2 models

Gemini 3.1 Pro Preview vs Gemini 3.7 Flash

Benchmarks through

Verdict

Gemini 3.7 Flash places higher on 9 of the 9 tasks where both are ranked. Gemini 3.7 Flash costs 2.7x less per input token.

  • Choose Gemini 3.7 Flash for Chat & Roleplay (#2), Translation (#2), Literary Translation (#2).

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 1Gemini 3.1 Pro Preview
Model 2Gemini 3.7 Flash
Best for
Business Writing (#3)
Business Writing (#2)
Provider
google
google
Context
1.0MBest
1.0MBest
Input / 1M
$2
$0.75Best
Output / 1M
$12
$3.75Best
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
—
—
Released
Feb 19, 2026
Aug 13, 2026

Measured head to head

Benchmarks both models were run on

EvaluationGemini 3.1 Pro PreviewGemini 3.7 FlashSource
Design Arena Coding1220 Elo1298 Elosource · 2026-11-15
Design Arena Full-stack Agent1041 Elo1183 Elosource · 2026-11-15
Design Arena UI1215 Elo1287 Elosource · 2026-11-15
Design Arena Web-app Agent1103 Elo1202 Elosource · 2026-11-15
Design Arena Website1225 Elo1295 Elosource · 2026-11-15
FrontierMath Tiers 1-3 (Epoch AI)59.6% ± 2.971.6% ± 2.7source · 2026-06-11
GPQA Diamond (Epoch AI)94.4% ± 1.694.8% ± 1.3source · 2026-08-06
GPQA Diamond (Epoch AI)94.1% ± 1.794.8% ± 1.3source · 2026-02-20
LiveBench Agentic Coding44.1%58.3%source · 2026-10-18
LiveBench Coding76.5%78.9%source · 2026-10-18
LiveBench Data Analysis78.5%68.0%source · 2026-10-18
LiveBench Instruction Following79.1%79.9%source · 2026-10-18
LiveBench Language85.4%85.5%source · 2026-10-18
LiveBench Math91.0%93.5%source · 2026-10-18
LiveBench Reasoning84%87.8%source · 2026-10-18
LMArena Agent-4.5-2.2source · 2026-11-11
LMArena Agent Bash Recovery-23.3-5.4source · 2026-10-08
LMArena Agent Steerability-8.2-7.3source · 2026-11-11
LMArena Agent Task Outcome-11.3-3.5source · 2026-11-11
LMArena Agent Tool Hallucination-0.40.7source · 2026-11-11
LMArena Creative Writing1483 Elo1485 Elosource · 2026-11-11
LMArena Instruction Following1465 Elo1485 Elosource · 2026-11-11
LMArena Long Query1482 Elo1494 Elosource · 2026-11-11
LMArena Math1485 Elo1506 Elosource · 2026-11-11
LMArena Text1487 Elo1489 Elosource · 2026-11-11
LMArena Text Factuality1472 Elo1476 Elosource · 2026-11-11
LMArena Vision1280 Elo1294 Elosource · 2026-11-09
LMArena Vision Diagrams1309 Elo1320 Elosource · 2026-11-09
LMArena Vision OCR1295 Elo1308 Elosource · 2026-11-09
LMArena WebDev1446 Elo1588 Elosource · 2026-11-11
LMArena WebDev Image-to-Web1480 Elo1600 Elosource · 2026-11-10
OTIS Mock AIME (Epoch AI)95.6% ± 3.197.2% ± 1.9source · 2026-08-06
OTIS Mock AIME (Epoch AI)95.6% ± 3.197.2% ± 1.9source · 2026-02-20
SimpleQA Verified (Epoch AI)73.5% ± 1.469.2% ± 1.5source · 2026-08-10

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