Vision-language

Gemini 1.5 Flash-8B

By Google DeepMind, in the Gemini Flash family.

Released
27 Aug 2024
Status
Retired
Retired
29 Sep 2025
Context
1M
Max output
8.2K
Input per 1M tokens
$0.04
Output per 1M tokens
$0.15
Open weights
No
License
proprietary
Parameters
8B
Knowledge cutoff
—
Reasoning
No
Modalities
text, image, audio, video → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Human preference21.9Not ranked2 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
DTBench (Decision Theory Benchmark)Accuracy (capability questions)50.0%Third party——https://conceptualreasoning.ai/dtbench via Epoch AI

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy33.0%Independent—12 Mar 2025Epoch AI

Mathematics

Mathematics results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
OTIS Mock AIME 2024-2025Accuracy4.6%Independent—25 Feb 2025Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Text Arena (arena.ai)Arena score (Elo scale)1259Independent—25 Sep 2026Arena
Text Arena (arena.ai)Votes35558Independent—25 Sep 2026Arena
Vision Arena (arena.ai)Arena score (Elo scale)1071Independent—28 Sep 2026Arena
Vision Arena (arena.ai)Votes6243Independent—28 Sep 2026Arena

Other Gemini Flash versions

Sources

Gemini 1.5 Flash-8B: benchmarks, price and specs · supermarket.dev