Skip to main content

Gemini 3.8 Flash vs Gemini 3 Flash Preview

SpecGemini 3.8 FlashGemini 3 Flash Preview
Respan Index54.6 (#30)41.0 (#65)
Blended price per 1M$1.5$1.13
Context window1.0M1.0M
Input / output per 1M$0.75 / $3.75$0.5 / $3
Max output66K66K
Inputtext, vision, audiotext, vision, audio
Released2026-09-022025-12-17

Respan Index by category

Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.

CategoryGemini 3.8 FlashGemini 3 Flash Preview
Coding36.1 Gemini 3.8 Flash11.8 Gemini 3 Flash Preview
Agents66.7 Gemini 3.8 Flash51.1 Gemini 3 Flash Preview
Reasoning53.3 Gemini 3.8 Flash42.8 Gemini 3 Flash Preview
Knowledge & language70.9 Gemini 3.8 Flash66.6 Gemini 3 Flash Preview
Human preference56.3 Gemini 3.8 Flash52.5 Gemini 3 Flash Preview

How the Respan Index works

Benchmarks both report

Gemini 3.8 Flash leads on 13 of the 14 benchmarks both report. Higher is better. Each score links to where it was published.

BenchmarkGemini 3.8 FlashGemini 3 Flash Preview
LMArena Elo14971473
LMArena WebDev (LMArena)15851440
LMArena Vision (LMArena)12931272
Chess Puzzles (Epoch AI)58%40%
FrontierMath Tier 4 v2 (Epoch AI)22%17.1%
FrontierMath Tiers 1-3 v2 (Epoch AI)68.4%51.2%
GPQA Diamond95.4%90.4%
Mystery Game Puzzles (Epoch AI)47%26%
OTIS Mock AIME 2024-2025 (Epoch AI)98.9%95.6%
SAGE (Vals AI)35.1%51.8%
SimpleQA Verified69.7%68.7%
Vending-Bench 2 (Andon Labs)5093.793634.72
WeirdML (Håvard Tveit Ihle)84.8%61.6%
SimpleBench (SimpleBench)82.4%61.1%

FAQ

Is Gemini 3.8 Flash better than Gemini 3 Flash Preview?

Gemini 3.8 Flash scores 54.6 on the Respan Index (#30), ahead of Gemini 3 Flash Preview at 41.0 (#65). Gemini 3.8 Flash leads on 13 of the 14 benchmarks both report.

Which is cheaper, Gemini 3.8 Flash or Gemini 3 Flash Preview?

Gemini 3.8 Flash lists at $0.75 input and $3.75 output per 1M tokens; Gemini 3 Flash Preview at $0.5 and $3. Gemini 3 Flash Preview is 1.3x cheaper per token at list price.

Which has the larger context window?

Both accept 1.0M tokens of context.

Compare other models

vs

Built for AI agents.
Break less.
Ship more.