Google model product
Gemma 3n E4B
Gemma 3n E4B is a Google multimodal language model designed to run locally on resource-constrained devices.
Updated Sep 8, 2026. Default version: Gemma 3n E4B Instructed
LLMBoard Score0.0Gemma 3n E4B Instructed
Coverage80%17 benchmark families
Context window32KTokens
Official input priceN/AOfficial price unavailable
Gemma 3n E4B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Gemma 3n E4B Benchmark Results
Benchmark scores for Gemma 3n E4B Instructed.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCodegolf v2.2 | Score16.80% | Rank01 | Participants4 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkECLeKTic | Score19.00% | Rank01 | Participants8 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkOpenAI MMLU | Score35.60% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGlobal-MMLU | Score60.30% | Rank02 | Participants5 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench v5 | Score25.70% | Rank06 | Participants9 | Percentile37.50% | EvidenceC | Evaluated |
| BenchmarkHiddenMath | Score37.70% | Rank08 | Participants13 | Percentile41.67% | EvidenceC | Evaluated |
| BenchmarkGlobal-MMLU-Lite | Score64.50% | Rank10 | Participants15 | Percentile35.71% | EvidenceC | Evaluated |
| BenchmarkWMT24++ | Score50.10% | Rank16 | Participants23 | Percentile31.82% | EvidenceC | Evaluated |
| BenchmarkMGSM | Score67.00% | Rank23 | Participants31 | Percentile26.67% | EvidenceC | Evaluated |
| BenchmarkInclude | Score57.20% | Rank26 | Participants31 | Percentile16.67% | EvidenceC | Evaluated |
| BenchmarkMBPP | Score63.60% | Rank26 | Participants33 | Percentile21.88% | EvidenceC | Evaluated |
| BenchmarkMMLU-ProX | Score19.90% | Rank29 | Participants32 | Percentile9.68% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score75.00% | Rank49 | Participants66 | Percentile26.15% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score13.20% | Rank72 | Participants75 | Percentile4.05% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score64.90% | Rank93 | Participants101 | Percentile8.00% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score11.60% | Rank117 | Participants119 | Percentile1.69% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score50.60% | Rank121 | Participants138 | Percentile12.41% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,305.23 rating | Rank159 | Participants210 | Percentile24.40% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,317.21 rating | Rank172 | Participants210 | Percentile18.18% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score23.70% | Rank244 | Participants247 | Percentile1.22% | EvidenceC | Evaluated |
Gemma 3n E4B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryjapanese | Rank120 | Rating / score1,273.84 | Votes723 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank120 | Rating / score1,292.51 | Votes723 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank130 | Rating / score1,311.82 | Votes720 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank130 | Rating / score1,259.52 | Votes513 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank134 | Rating / score1,273.44 | Votes513 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank135 | Rating / score1,319.81 | Votes720 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank136 | Rating / score1,331.30 | Votes326 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank137 | Rating / score1,285.55 | Votes1,915 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank138 | Rating / score1,304.48 | Votes451 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank140 | Rating / score1,346.13 | Votes326 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank141 | Rating / score1,304.86 | Votes1,915 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank153 | Rating / score1,308.85 | Votes1,490 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank153 | Rating / score1,321.15 | Votes1,321 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank154 | Rating / score1,309.61 | Votes451 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank155 | Rating / score1,286.90 | Votes2,848 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank155 | Rating / score1,323.76 | Votes3,669 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank156 | Rating / score1,294.98 | Votes3,268 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank157 | Rating / score1,284.46 | Votes12,013 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank158 | Rating / score1,308.86 | Votes1,386 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank158 | Rating / score1,287.04 | Votes1,644 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank159 | Rating / score1,305.23 | Votes22,087 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank159 | Rating / score1,330.87 | Votes1,490 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank159 | Rating / score1,298.71 | Votes2,848 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank160 | Rating / score1,321.31 | Votes10,068 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank160 | Rating / score1,244.54 | Votes15,678 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank161 | Rating / score1,307.61 | Votes6,408 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank161 | Rating / score1,286.38 | Votes4,850 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank163 | Rating / score1,283.36 | Votes8,464 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank163 | Rating / score1,261.95 | Votes3,868 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry business and management and financial operations | Rank165 | Rating / score1,316.33 | Votes3,268 | ObservationsN/A | Result date |
Gemma 3n E4B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Gemma 3n E4B Runtime Performance
Provider-specific output speed and catalog latency for Gemma 3n E4B Instructed. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Gemma 3n E4B Specifications
Technical details for the model's default version.
- Version
- Gemma 3n E4B Instructed
- Released
- Jun 26, 2025
- Knowledge cutoff
- Jun 1, 2024
- Parameters
- 8B
- Context window
- 32K
- Max output
- 32K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Proprietary
Gemma 3n E4B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemma 3n E4B | Released | LLMBoardN/A | Parameters8B | ContextN/A | Max outputN/A | Open weightsYes | LicenseProprietary |
| VersionGemma 3n E4B Instructed | Released | LLMBoard0.00 | Parameters8B | Context32K | Max output32K | Open weightsYes | LicenseProprietary |
Models similar to Gemma 3n E4B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
GO
Google
MedGemma 4B IT
GOGoogle
Gemini 1.0 Pro
GOGoogle
Gemini 1.5 Flash 8B
GOGoogle
Gemma 2 27B
GOGoogle
Gemma 2 9B
GOGoogle
Gemma 3 1B
What is Gemma 3n E4B?
Key information about Gemma 3n E4B and its available data.
Gemma 3n E4B is an E4B-sized model in Google’s Gemma 3n family, accepting text, image, audio, and video inputs and generating text outputs. It includes language decoder, audio encoder, and vision encoder components and is optimized for memory efficiency.
Data as of 2026-09-08.
FAQ
Common questions about Gemma 3n E4B.
When was Gemma 3n E4B released?
Gemma 3n E4B's default version was released on Jun 26, 2025.
How much does Gemma 3n E4B cost?
No official standard PAYG price is currently available for Gemma 3n E4B.
Who created Gemma 3n E4B?
Gemma 3n E4B was created by Google.
What is the context window for Gemma 3n E4B?
The default version has a 32K token context window.
Is Gemma 3n E4B open weight?
Yes. The default version is marked as open weight under Proprietary.
How many API providers offer Gemma 3n E4B?
No provider offering is currently linked to the default version.