Gemma 3 4B
Google's compact open model that runs on most consumer hardware while retaining solid instruction quality.
Pricing
InputOpen weights
OutputSelf-hosted cost profile
Cost bandFree/open
Free tierOpen weights.
Limits
Context window128k checked Jul 19, 2026
Max outputRuntime-dependent
Knowledge cutoffFeb 2025
DeploymentHugging Face, Ollama, self-hosted
OpenAI-compatible APIYes
ModalitiesText, Image input
Where it fits
Best forEdge deployments, on-device inference, and scenarios where memory footprint matters most.
Business fitBest open model for ultra-constrained deployment environments.
Capabilities
Open weightsEdge inferenceMultilingual
Benchmark scores
Composite55 / 100
Coding52 / 100 (HumanEval)
Reasoning54 / 100 (MMLU-Pro)
Long context50 / 100
Vision60 / 100 (MMMU)
Instruction following58 / 100 (IFEval)
Editorial aggregate of published benchmark results as of Apr 1, 2025. These are not independent measurements run by this site.
Change history
No recorded changes for this model yet. Pricing and context are checked daily; changes appear here with dates and sources.
Data
Last checkedJul 19, 2026
Cross-checked againstopenrouter.ai/google/gemma-3-4b-it
Verify with provider documentation before committing spend. Report an error and it gets fixed with a logged correction.