Gemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.
Context Window
128.0K
Max Output
8.2K
Avg output tokens (7d)
236 tokens
Input Price (Auto)
$0.20/1M
Output Price (Auto)
$0.20/1M
Cache Read (Auto)
$0.10/1M
Capabilities
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
1.0
Coding Index
2.7
Reasoning
GPQA Diamond
Graduate-level scientific reasoning
29.1%
Better than 5% of models compared
HLE
Humanity's Last Exam
5.3%
Better than 34% of models compared
IFBench
Instruction-following benchmark
28.3%
Better than 12% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
5.0%
Better than 6% of models compared
AA-LCR
Long context reasoning evaluation
6.7%
Better than 14% of models compared
Coding
Terminal-Bench Hard
Agentic coding and terminal use
0.8%
Better than 13% of models compared
LiveCodeBench
Contamination-free coding benchmark
11.2%
Better than 9% of models compared
Math
AIME 2025
American Invitational Mathematics Examination 2025
12.7%
Better than 15% of models compared
AIME
American Invitational Mathematics Examination
6.3%
Better than 22% of models compared
Math-500
Diverse mathematical problem solving benchmark
76.6%
Better than 39% of models compared
Knowledge
MMLU-Pro
Professional and academic subject knowledge
41.7%
Better than 9% of models compared
Last updated Sep 7, 2026
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Related text models
Compare Gemma 3 4B IT with similar models from the same provider or model family.
Gemma 3 12B IT
unsloth/gemma-3-12b-itGemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.
Gemma 3 27B IT
unsloth/gemma-3-27b-itGemma 3 has a large, 128K context window, multilingual support in over 140 languages, and is available in more sizes than previous versions.
Gemini 3.8 Flash
google/gemini-3.8-flashGoogle's fast multimodal model for agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video. Its capabilities, limits, reasoning behavior, and pricing currently mirror Gemini 3.7 Flash.
Gemini 3.7 Flash
google/gemini-3.7-flashGoogle's frontier-performance Flash model for multimodal and agentic workloads, including coding, tool use, image understanding, PDF and document extraction, audio, and video. Google reports 65.3% on DeepSWE v1.1, up from 49.0% for Gemini 3.6 Flash, and 34% on GDP.pdf, up from 14%.
Gemini 3.5 Flash Lite
google/gemini-3.5-flash-liteGoogle's cost-efficient Gemini 3.5 Flash Lite model for high-volume multimodal reasoning, tool use, and structured-output workloads.
Gemini 3.6 Flash
google/gemini-3.6-flashGoogle's GA Flash model delivers near-Pro coding and agentic capability at Flash-tier speed and cost, with improved token efficiency over Gemini 3.5 Flash. It is optimized for multi-step orchestration, full-stack code refactoring, general reasoning, and multimodal workflows.