SAVRN
Search Contact SAVRN

Open-weight model · Any to any

gemma-4-E4B-it

by Google google/gemma-4-E4B-it

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output.

Parameters8B
Context131,072
Weights16.0 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads4.5M

Runs On

What it takes to serve gemma-4-E4B-it (8B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 16.0 GB 19.2 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 8.0 GB 9.6 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 4.0 GB 4.8 GB 1x MI300X (192 GB)
Vultr		 $1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

SAVRN's Notes on gemma-4-E4B-it

The name says E4B, the file counts 8 billion parameters, and memory follows the second number: 16.0 GB of weights at 16-bit and 19.2 GB needed. That fits the cheapest setup on our Index, one 192 GB MI300X at $1.85 an hour on-demand, with room for nine more copies. You get an instruction-tuned model taking text, images and audio in and writing text out, so one deployment covers several channels.

On licensing, Apache 2.0 covers commercial use, modification and redistribution, provided the license notices and a statement of significant changes travel with it. Check the context: the publisher describes the family at up to 256K tokens, this checkpoint's configuration says 131,072, and 131,072 is what you deploy against. It derives from google/gemma-4-E4B; start there if you plan your own tuning. The scores come from the model card and outside evaluators, none from us, and no Index host prices it.

SAVRN Research, 2026-09-18

Model Card

By Google, published under apache-2.0, revision ee0ef6023621.

Hugging Face | GitHub | Launch Blog | Documentation | Technical Report
License: Apache 2.0 | Authors: Google DeepMind

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages.

Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from high-end phones to laptops and servers, democratizing access to state-of-the-art AI.

Gemma 4 introduces key capability and architectural advancements:

Read the full model card (3,425 words)

Configuration

Architecture
Gemma4ForConditionalGeneration
Context length (tokens)
131,072
Layers
42
Hidden size
2,560
Feed-forward size
10,240
Attention heads
8
Key/value heads
2
Head dimension
256
Vocabulary size
262,144
Sliding window (tokens)
512
Model type
gemma4

Identity and Version

Repository
google/gemma-4-E4B-it
Publisher
Google
Task
Any to any
Modality
Multimodal
Library
transformers
Parameters
8B parameters
Languages
Not stated by the source
Revision
ee0ef6023621cff504d758262d4e04895a5af4a2
First published
2026-03-02
Last updated
2026-07-20

Files and Weights

9 files, 16.0 GB in total. The weights are 1 file totalling 16.0 GB in safetensors.

Weights1 file · 16.0 GB
Configuration3 files · 7.0 KB
Tokenizer2 files · 32.2 MB
Documentation1 file · 28.0 KB
Other1 file · 18.6 KB
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights16.0 GB cfbd3d2f1cd7
config.jsonConfiguration5.1 KB —
generation_config.jsonConfiguration208 B —
processor_config.jsonConfiguration1.7 KB —
README.mdDocumentation28.0 KB —
chat_template.jinjaOther18.6 KB —
.gitattributesRepository1.6 KB —
tokenizer.jsonTokenizer32.2 MB cc8d3a0ce364
tokenizer_config.jsonTokenizer3.1 KB —

License and Download

License
apache-2.0
Access
Open weights, no gate
Download size
16.0 GB
Download from Google

Released by Google through Kaggle. Read the license.

Built From

Evaluations

Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.

BenchmarkConditionsResultReported byRevisionDate
Idavidrein/gpqa Task diamondMetric diamondComparison conditions not established 58.6 Model Card
Reported by a third party		 Evaluated revision not stated 2026-04-02
LiquidAI/ifstruct-v1.0 Task ifstruct_v1Metric ifstruct_v1Comparison conditions not established 76.65 Liquid AI — IFStruct v1.0 blog (gemma-4-E4B-it)
Reported by a third party		 Evaluated revision not stated 2026-06-30
MMMU/MMMU_Pro Task mmmu_pro_visionMetric mmmu_pro_visionComparison conditions not established 52.6 Model Card
Reported by a third party		 Evaluated revision not stated 2026-05-12
MathArena/aime_2026 Task MathArena/aime_2026Metric MathArena/aime_2026Comparison conditions not established 42.5 Model Card
Reported by a third party		 Evaluated revision not stated 2026-03-02
TIGER-Lab/MMLU-Pro Task mmlu_proMetric mmlu_proComparison conditions not established 69.4 Model Card
Reported by a third party		 Evaluated revision not stated 2026-04-02
llamaindex/ExtractBench Task longMetric longSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 26.82 ExtractBench
Reported by a third party		 Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task meanMetric meanSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 69.64 ExtractBench
Reported by a third party		 Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task mediumMetric mediumSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 48.93 ExtractBench
Reported by a third party		 Evaluated revision not stated 2026-08-24
llamaindex/ExtractBench Task shortMetric shortSetup Pipeline name: gemma4_e4b_vllm_extract_oneshot_structured_output_fileComparison conditions not established 81.1 ExtractBench
Reported by a third party		 Evaluated revision not stated 2026-08-24
llamaindex/ParseBench Task chartMetric chartSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 7.7 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task layoutMetric layoutSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 35.8 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task meanMetric meanSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 40.5 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task tableMetric tableSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 25 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task text_contentMetric text_contentSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 81.4 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14
llamaindex/ParseBench Task text_formattingMetric text_formattingSetup Pipeline name: gemma4_e4b_vllm_with_layoutComparison conditions not established 52.5 ParseBench
Reported by a third party		 Evaluated revision not stated 2026-04-14

Memory Requirements

PrecisionWeights in memory
As published16.0 GB
16-bit16.0 GB
8-bit8.0 GB
4-bit4.0 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Built on This Model

Compare gemma-4-E4B-it

Questions About gemma-4-E4B-it

How much GPU memory does gemma-4-E4B-it need?

About 19.2 GB at 16-bit and 4.8 GB at 4-bit: the weights (8B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run gemma-4-E4B-it on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use gemma-4-E4B-it commercially?

Yes. gemma-4-E4B-it is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is gemma-4-E4B-it's context length?

131,072 tokens, from the maximum position embeddings in its published configuration.

Similar Models

LM Studio Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on Discord. 4-bit quantized version of gemma-4-E4B-it using MLX, optimized for Apple Silicon. Special thanks to the Apple Machine Learning Research team for creating MLX. LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model. You understand that Community Models can produce content that might be offensive, harmful…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model

LM Studio Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on Discord. 8-bit quantized version of gemma-4-E4B-it using MLX, optimized for Apple Silicon. Special thanks to the Apple Machine Learning Research team for creating MLX. LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model. You understand that Community Models can produce content that might be offensive, harmful…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model

LM Studio Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on Discord. 5-bit quantized version of gemma-4-E4B-it using MLX, optimized for Apple Silicon. Special thanks to the Apple Machine Learning Research team for creating MLX. LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model. You understand that Community Models can produce content that might be offensive, harmful…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model

LM Studio Community models highlights program. Highlighting new & noteworthy models by the community. Join the conversation on Discord. 6-bit quantized version of gemma-4-E4B-it using MLX, optimized for Apple Silicon. Special thanks to the Apple Machine Learning Research team for creating MLX. LM Studio is not the creator, originator, or owner of any Model featured in the Community Model Program. Each Community Model is created and provided by third parties. LM Studio does not endorse, support, represent or guarantee the completeness, truthfulness, accuracy, or reliability of any Community Model. You understand that Community Models can produce content that might be offensive, harmful…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model

Model · Any to any

gemma-4-E4B

Google

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages. Featuring both Dense and Mixture-of-Experts (MoE) architectures, Gemma 4 is well-suited for tasks like text generation, coding, and reasoning. The models are available in five distinct sizes: E2B, E4B, 12B, 26B A4B, and 31B. Their diverse sizes make them deployable in environments ranging from…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model

Creating these models takes significant time, work and compute. If you find them useful consider supporting me: Your help will motivate me and would go into further improving my workflow and coverings fees for storage, compute and may even help uncensoring bigger model with rental Cloud GPUs. attn.oproj Lower refusals indicate fewer content restrictions, while lower KL divergence indicates more closeness to the original model's baseline. Higher refusals cause more rejections, objections, pushbacks, lecturing, censorship, softening and deflections. PIQA (Physical Intuition Question Answering) a ~1,800 questions tests common-sense understanding of how the physical world works with benchmark…

Open weights apache-2.0 8B parameters 131,072 tokens transformers
View model