Google DeepMind

Gemini 3 Flash

Multimodal

by Google DeepMind

+
+
+
+
About

Gemini 3 Flash, released by Google in December 2025, brings the reasoning and coding improvements of the Gemini 3 generation to a faster, more cost-efficient model. It targets high-throughput applications that need Gemini 3's multimodal quality at a more accessible price point, continuing Google's pattern of releasing efficient Flash variants alongside each major model generation.

+
+
+
+
Pricing Range
Input (per 1M)$0.10 -$0.10
Output (per 1M)$0.40 -$0.40
Providers2
+
+
+
+
Timeline
ReleasedDec 17, 2025
+
+
+
+
Specifications
Capabilities
Multimodal
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown

Overall Performance

4 benchmarks
Average Score
55.6%
Best Score
78.0%
High Performers (80%+)
0

Performance Metrics

Max Context Window
1.0M

Top Categories

Coding
63.0%
Reasoning
33.6%
+
+
+
+
All Benchmark Results for Gemini 3 Flash
Complete list of benchmark scores with detailed information
SWE Bench Verified
Coding
78.00
78.0%
Unverified
Terminal Bench 2.0
Coding
64.30
64.3%
Unverified
SWE-rebench
Coding
46.70
46.7%
Unverified
ARC-AGI-2
Reasoning
33.60
33.6%
Unverified