Comparing 2 models

Gemini 3.5 Flash vs Gemini 3.8 Flash

Benchmarks through

Verdict

Gemini 3.8 Flash places higher on 11 of the 11 tasks where both are ranked. Gemini 3.8 Flash costs 1.8x less per input token.

  • Choose Gemini 3.8 Flash for Chat & Roleplay (#1), Translation (#1), Literary Translation (#1).

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 3.5 Flash
Model 2Gemini 3.8 Flash
Best for
Business Writing (#5)
Business Writing (#1)
Provider
google
google
Context
1.0MBest
1.0MBest
Input / 1M
$1.50
$0.83Best
Output / 1M
$9
$4.13Best
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
—
—
Released
May 19, 2026
Sep 2, 2026

Measured head to head

Benchmarks both models were run on

EvaluationGemini 3.5 FlashGemini 3.8 FlashSource
Design Arena Coding1256 Elo1292 Elosource · 2026-11-15
Design Arena Full-stack Agent1170 Elo1216 Elosource · 2026-11-15
Design Arena UI1265 Elo1303 Elosource · 2026-11-15
Design Arena Web-app Agent1177 Elo1213 Elosource · 2026-11-15
Design Arena Website1260 Elo1291 Elosource · 2026-11-15
FrontierMath Tiers 1-3 (Epoch AI)62.8% ± 2.968.4% ± 2.8source · 2026-06-10
GPQA Diamond (Epoch AI)88.9% ± 2.295.4% ± 1.4source · 2026-08-06
GPQA Diamond (Epoch AI)86.4% ± 2.495.4% ± 1.4source · 2026-07-15
GPQA Diamond (Epoch AI)92.8% ± 1.695.4% ± 1.4source · 2026-05-22
LiveBench Agentic Coding49.0%54.2%source · 2026-10-18
LiveBench Coding78.2%72.5%source · 2026-10-18
LiveBench Data Analysis64.9%54.0%source · 2026-10-18
LiveBench Instruction Following75.6%81.4%source · 2026-10-18
LiveBench Language84.6%87.8%source · 2026-10-18
LiveBench Math88.2%91.6%source · 2026-10-18
LiveBench Reasoning82%89.3%source · 2026-10-18
LMArena Creative Writing1469 Elo1493 Elosource · 2026-11-11
LMArena Instruction Following1467 Elo1493 Elosource · 2026-11-11
LMArena Long Query1484 Elo1508 Elosource · 2026-11-11
LMArena Math1504 Elo1525 Elosource · 2026-11-11
LMArena Text1478 Elo1497 Elosource · 2026-11-11
LMArena Text Factuality1478 Elo1470 Elosource · 2026-11-11
LMArena Vision1281 Elo1293 Elosource · 2026-11-09
LMArena Vision Diagrams1307 Elo1342 Elosource · 2026-11-09
LMArena Vision OCR1291 Elo1315 Elosource · 2026-11-09
LMArena WebDev1500 Elo1583 Elosource · 2026-11-11
LMArena WebDev Image-to-Web1500 Elo1594 Elosource · 2026-11-10
OTIS Mock AIME (Epoch AI)88.9% ± 4.798.9% ± 0.9source · 2026-08-06
OTIS Mock AIME (Epoch AI)80% ± 698.9% ± 0.9source · 2026-07-15
OTIS Mock AIME (Epoch AI)95.6% ± 2.798.9% ± 0.9source · 2026-05-25
SimpleQA Verified (Epoch AI)66.2% ± 1.569.7% ± 1.5source · 2026-08-27

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