AI BENCHY
Advertise here
#1

Gemini 3 Flash Preview

Google Release: 2025-12-17 Tested on: 2026-07-16 22:51 google/gemini-3-flash-preview::medium
~500B total (~20B active)MoEClosedEstimated
(medium) (low) (none)

Summary

Gemini 3 Flash Preview scores 9.6 on AI BENCHY and ranks #1. It has 10.0 reliability, a 98.5% pass rate, $0.742 total cost, and 19.20s average response time.

What makes Gemini 3 Flash Preview unique: It stands out most in Domain specific, where it ranks #1, while Coding is its weakest area at #8. It is one of the highest-ranked models overall.

Model facts

Researched on 2026-08-12

Estimated
Parameters
~500B total (~20B active)
Architecture
MoE
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Sparse MoE is documented for the family; counts are estimated.

Parameter source Availability source

Score

9.6

Consistency

9.7

Reliability

10.0

Total Cost (Current Price)

$0.742

Total Output Tokens

232,650

Total Input Tokens

87,861

Input Price

$0.500 / 1M

Output Price

$3.000 / 1M

Tests Correct

Wrong Tests: 1

Attempt pass rate: 98.5%

Flaky tests

1

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

19.20s

Response Time (max): 117.26s

Response Time (total): 422.42s

Wrong answer: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#1 Gemini 3 Flash Preview

medium
Cost
$0.010
Time
18.4s
Tokens
3,351 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-08-14 02:41 Re-test 9.5 10.0 $0.763 Compare
2026-07-16 22:51 New test added 9.6 10.0 $0.742 Current run
2026-06-04 13:59 New test added 9.8 10.0 $0.667 Compare
2026-05-22 00:28 Suite changed 9.8 10.0 $0.567 Compare
2026-05-06 13:52 Suite changed 10.0 10.0 $0.321 Compare
2026-05-06 13:38 Suite changed 9.8 10.0 $0.320 Compare
2026-04-23 15:25 Re-test 10.0 N/A $0.108 Compare
2026-04-11 01:44 First recorded run 10.0 N/A $0.314 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-07-16 22:51 · New test added66/66 attempts9.69.710.021/221232,65087,861$0.74219.20s
2026-05-06 13:38 · Suite changed54/54 attempts9.89.610.017/181101,4800$0.32011.53s
Difference—-0.2+0.10.0+40+131170+87861+$0.423+7675ms

Benchmark coverage differs: 66/66 attempts (Target: 3 repeats per test) versus 54/54 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Price History

Historical pricing data for this model from OpenRouter.

Date Input Price Output Price
2026-06-04 15:40 $0.500 / 1M $3.000 / 1M

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Gemini 3 Flash PreviewmediumvsGPT-5.6 SollowGemini 3 Flash PreviewmediumvsGemini 3.5 FlashhighGemini 3 Flash PreviewmediumvsGemini 3.6 FlashmediumGemini 3 Flash PreviewmediumvsGemini 3.7 FlashmediumGemini 3 Flash PreviewmediumvsGemini 3.7 FlashhighGemini 3 Flash PreviewmediumvsGPT-5.6 SolmediumGemini 3 Flash PreviewmediumvsGemini 3.6 FlashhighGemini 3 Flash PreviewmediumvsGPT-5.6 SolhighGemini 3 Flash PreviewmediumvsGPT-5.5lowGemini 3 Flash PreviewmediumvsGrok 4.6high

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 10.0 10.0
Coding 8.6 7.6
Combined 10.0 10.0
Data parsing and extraction 10.0 10.0
Domain specific 10.0 10.0
General Intelligence 10.0 10.0
Instructions following 10.0 10.0
Puzzle Solving 10.0 10.0
Tool Calling 10.0 10.0
Trivia 10.0 10.0

Compared models