Skip to content
Leaderboard / Google

Gemini 3.5 Flash

Google High CLOSED Proprietary
Other settings:Medium · 83
Verdict

Gemini 3.5 Flash is a strong all-rounder: #29 of 406 by SIscope score. Strongest at chat quality, weaker at coding. Mid-range on price.

SIscope score
85
#29 of 406
Human-vote rank
#19
of 419 · score 1482
Price per 1M tokens, blended
$3.38
$1.5 in · $9 out
Context window
1M
1,048,576 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsno data yetReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding81
Agents & toolsNo data yet–
Reasoning & knowledge92
Math82
Chat quality96
Following instructions92

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

147514831490Oct 20: 1482 (rank #18)Oct 22: 1481.9 (rank #18)Oct 28: 1482.2 (rank #19)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

149515001505Oct 16: 1499.6 (rank #59)Oct 18: 1499.5 (rank #60)Oct 19: 1499.4 (rank #62)Oct 22: 1499.2 (rank #62)Oct 23: 1499.2 (rank #63)Oct 27: 1499.1 (rank #64)Oct 28: 1499.3 (rank #65)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond93%
PhD-level science multiple choice questions
Better than 93% of 333 tested models. Run May 22, 2026.
SimpleQA Verified66%
Short factual questions; rewards not making things up
Better than 88% of 100 tested models. Run Aug 27, 2026.
OTIS Mock AIME 2024-2596%
Competition maths problems in the style of AIME
Better than 82% of 313 tested models. Run May 25, 2026.
FrontierMath Tiers 1-363%
Very hard, unpublished research-style maths problems
Better than 67% of 122 tested models. Run Jun 10, 2026.
SWE-bench Verified79%
Fixing real GitHub issues in Python repositories
Better than 94% of 35 tested models. Run Jun 1, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#19 of 41914821478–148648,376Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#65 of 15214991492–15079,331Oct 28, 2026
Vision arena
Questions about images, blind human votes
#13 of 16713081302–131513,862Oct 19, 2026
Agent arena
Real multi-step agent sessions
Not on this board yet

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$1.5
per 1M tokens
Output
$9
per 1M tokens
Blended
$3.38
3 in : 1 out, per 1M
Context window
1M
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (google/gemini-3.5-flash), Nov 4, 2026. Providers can charge differently.