Skip to content
English

Switch theme

Google

Gemini 3.8 Flash benchmark results

7 results on 7 benchmarks from 3 sources.

Closed Released: Sep 2, 2026 Updated Oct 22, 2026 API prices & specs →

Also listed as: gemini-3.8-flash-high · gemini-3.8-flash_high · Gemini 3.8 Flash (high) · gemini-3.8-flash_unknown · Gemini 3.8 Flash (unknown) · Gemini 3.8 Flash (High)

All benchmark results

Benchmarks Score Rank Setting Source type Date
LMArena Text 1497 8 Overall, style control on; 31437 votes; 95% CI 1492–1501“gemini-3.8-flash-high” Official leaderboard Oct 8, 2026
LMArena WebDev 1584 31 Overall; 9225 votes; 95% CI 1576–1591“gemini-3.8-flash-high” Official leaderboard Oct 8, 2026
LMArena Vision 1290 11 Overall, style control on; 3086 votes; 95% CI 1279–1302“gemini-3.8-flash-high” Official leaderboard Oct 2, 2026
GPQA Diamond 95.4% 4 Epoch AI run, accuracy; effort: high Independent run Sep 2, 2026
AIME (OTIS Mock AIME 2024–2025) 98.9% 20 Epoch AI run, OTIS Mock AIME 2024–2025; effort: high Independent run Sep 2, 2026
Humanity's Last Exam 44.5% 5 Compiled by Epoch AI; effort: unknown Third-party compilation via Epoch AI Sep 2, 2026
ARC-AGI-1 98.5% 6 Semi-private set; $0.21/task; CoT“Gemini 3.8 Flash (High)” Official leaderboard Oct 22, 2026

Related on this site

← AI Model Benchmarks

FAQ

Where do these Gemini 3.8 Flash scores come from?

From Epoch AI Benchmarking Hub, ARC Prize leaderboard, and LMArena leaderboard dataset. Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.

Data updated: Benchmarks(UTC+8)