# embeddinggemma-300m by Google: Open-Weight Model Source: https://savrn.com/models/embeddinggemma-300m Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve embeddinggemma-300m (303M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.6 GB | 0.7 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.3 GB | 0.4 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.2 GB | 0.2 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026. [embeddinggemma-300m on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/embeddinggemma-300m/gpus) ## SAVRN's Notes on embeddinggemma-300m Retrieval is where this one earns its place: every document indexed and every query served passes through it, and at 0.7 GB of memory at 16-bit, EmbeddingGemma-300m never justifies a GPU on its own. Google built it from Gemma 3 and aimed it at phones, laptops and desktops, so share a card with the generator, or skip the GPU. The Index floor, one MI300X with 192 GB at $1.85 per hour, is a card you would barely touch. Read the license twice. Gemma Terms of Use allow commercial use and redistribution only subject to the Gemma Prohibited Use Policy, and those restrictions pass to anyone you distribute it to, a flow-down obligation for any product built on it. Access is gated; the publisher grants it before the 19 files arrive. No context length is listed, so test your chunk size against the checkpoint dated September 25, 2025. SAVRN Research, 2026-09-18 ## Model Card EmbeddingGemma is a 300M parameter, state-of-the-art for its size, open embedding model from Google, built from Gemma 3 (with T5Gemma initialization) and the same research and technology used to create Gemini models. EmbeddingGemma produces vector representations of text, making it well-suited for search and retrieval tasks, including classification, clustering, and semantic similarity search. This model was trained with data in 100+ spoken languages. The small size and on-device focus makes it possible to deploy in environments with limited resources such as mobile phones, laptops, or desktops, democratizing access to state of the art AI models and helping foster innovation for everyone.… Excerpt from the card by Google, licensed gemma. ## Identity and Version Repository google/embeddinggemma-300m Publisher Google Task Sentence similarity Modality Text Library sentence-transformers Parameters 303M parameters Languages Not stated by the source Revision 57c266a740f537b4dc058e1b0cda161fd15afa75 First published 2025-07-17 Last updated 2025-09-25 ## Files and Weights 19 files, 1.3 GB in total. The weights are 3 files totalling 1.2 GB in safetensors. Weights3 files · 1.2 GB Configuration10 files · 4.5 KB Tokenizer3 files · 39.2 MB Documentation1 file · 18.7 KB Other1 file · 28.8 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | 2_Dense/model.safetensors | Weights | 9.4 MB | — | | 3_Dense/model.safetensors | Weights | 9.4 MB | — | | model.safetensors | Weights | 1.2 GB | — | | 1_Pooling/config.json | Configuration | 312 B | — | | 2_Dense/config.json | Configuration | 134 B | — | | 3_Dense/config.json | Configuration | 134 B | — | | added_tokens.json | Configuration | 35 B | — | | config.json | Configuration | 1.5 KB | — | | config_sentence_transformers.json | Configuration | 997 B | — | | generation_config.json | Configuration | 133 B | — | | modules.json | Configuration | 573 B | — | | sentence_bert_config.json | Configuration | 58 B | — | | special_tokens_map.json | Configuration | 662 B | — | | README.md | Documentation | 18.7 KB | — | | notebook.ipynb | Other | 28.8 KB | — | | .gitattributes | Repository | 1.6 KB | — | | tokenizer.json | Tokenizer | 33.4 MB | — | | tokenizer.model | Tokenizer | 4.7 MB | — | | tokenizer_config.json | Tokenizer | 1.2 MB | — | ## License and Download License gemma Access Access requested at publisher Download size 1.2 GB [Download from Google](https://www.kaggle.com/models/google/embeddinggemma) Released by Google through Kaggle. ## Built From - Described by arXiv:2509.20354 ## 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. | Benchmark | Conditions | Result | Reported by | Revision | Date | | --- | --- | --- | --- | --- | --- | | mteb/arguana | Task ArguAnaMetric ArguAnaSetup Obtained using MTEB v1.34.7Comparison conditions not established | 71.535 | [Obtained using MTEB v1.34.7](https://github.com/embeddings-benchmark/mteb/) Reported by a third party | Evaluated revision not stated | 2026-03-05 | | mteb/arguana | Task ArguAna_default_testMetric ArguAna_default_testSetup Obtained using MTEB v1.34.7Comparison conditions not established | 71.535 | [Obtained using MTEB v1.34.7](https://github.com/embeddings-benchmark/mteb/) Reported by a third party | Evaluated revision not stated | 2026-03-05 | ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 1.2 GB | | 16-bit | 0.6 GB | | 8-bit | 0.3 GB | | 4-bit | 0.2 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Built on This Model - Derived from[universal_classifier_nouls](https://savrn.com/models/universal-classifier-nouls) - Quantized from[embeddinggemma-300m-memory-ft-v2](https://savrn.com/models/embeddinggemma-300m-memory-ft-v2) - Derived from[embeddinggemma-300m-memory-ft-v2](https://savrn.com/models/embeddinggemma-300m-memory-ft-v2) ## Compare embeddinggemma-300m - [embeddinggemma-300m vs gte-multilingual-base](https://savrn.com/models/compare/embeddinggemma-300m-vs-gte-multilingual-base) - [embeddinggemma-300m vs multilingual-e5-base](https://savrn.com/models/compare/embeddinggemma-300m-vs-multilingual-e5-base) - [embeddinggemma-300m vs paraphrase-multilingual-mpnet-base-v2](https://savrn.com/models/compare/embeddinggemma-300m-vs-paraphrase-multilingual-mpnet-base-v2) ## Questions About embeddinggemma-300m ### How much GPU memory does embeddinggemma-300m need? About 0.7 GB at 16-bit and 0.2 GB at 4-bit: the weights (303M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run embeddinggemma-300m 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 embeddinggemma-300m commercially? Yes, with conditions. embeddinggemma-300m is released under Gemma Terms of Use. Gemma models are released under Google's Gemma Terms of Use, which permit commercial use and redistribution subject to the Gemma Prohibited Use Policy, whose restrictions must be passed on to anyone the model is distributed to. ## Similar Models Model · Sentence similarity ### [gte-multilingual-base](https://savrn.com/models/gte-multilingual-base) [Alibaba-NLP](https://savrn.com/model-publishers/alibaba-nlp) The gte-multilingual-base model is the latest in the GTE (General Text Embedding) family of models, featuring several key attributes: Paper: mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval - It is recommended to install xformers and enable unpadding for acceleration, refer to enable-unpadding-and-xformers. - How to use with TEI: refs/pr/7 Usage via docker and infinity, MIT Licensed. Usage via Docker and Text Embeddings Inference (TEI): Then you can send requests to the deployed API via the OpenAI-compatible v1/embeddings route (more information about the OpenAI Embeddings API): We validated the performance of the gte-multilingual-base… Open weights apache-2.0 305M parameters 8,192 tokens sentence-transformers [View model](https://savrn.com/models/gte-multilingual-base) Model · Sentence similarity ### [paraphrase-multilingual-mpnet-base-v2](https://savrn.com/models/paraphrase-multilingual-mpnet-base-v2) [Sentence Transformers](https://savrn.com/model-publishers/sentence-transformers) This is a sentence-transformers model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Using this model becomes easy when you have sentence-transformers installed: Then you can use the model like this: Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings. Text Embeddings Inference (TEI) is a blazing fast inference solution for text embedding models. Send a request to /v1/embeddings to generate embeddings via the OpenAI Embeddings API: Or check… Open weights apache-2.0 278M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/paraphrase-multilingual-mpnet-base-v2) Model · Sentence similarity ### [multilingual-e5-base](https://savrn.com/models/multilingual-e5-base) [Liang Wang](https://savrn.com/model-publishers/intfloat) Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei, arXiv 2024 This model has 12 layers and the embedding size is 768. Below is an example to encode queries and passages from the MS-MARCO passage ranking dataset. This model is initialized from xlm-roberta-base and continually trained on a mixture of multilingual datasets. It supports 100 languages from xlm-roberta, but low-resource languages may see performance degradation. For all labeled datasets, we only use its training set for fine-tuning. For other training details, please refer to our paper at https://arxiv.org/pdf/2402.05672. Check out unilm/e5 to reproduce evaluation results on the BEIR and MTEB benchmark.… Open weights mit 278M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/multilingual-e5-base) Model · Sentence similarity ### [lt-wikidata-comp-multi](https://savrn.com/models/lt-wikidata-comp-multi) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 278M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-wikidata-comp-multi) Model · Sentence similarity ### [lt-un-data-fine-industry-multi](https://savrn.com/models/lt-un-data-fine-industry-multi) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 278M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-un-data-fine-industry-multi) Model · Sentence similarity ### [lt-un-data-fine-fine-multi](https://savrn.com/models/lt-un-data-fine-fine-multi) [Dell Research Harvard](https://savrn.com/model-publishers/dell-research-harvard) This is a LinkTransformer model. At its core this model this is a sentence transformer model sentence-transformers model- it just wraps around the class. It is designed for quick and easy record linkage (entity-matching) through the LinkTransformer package. The tasks include clustering, deduplication, linking, aggregation and more. Notwithstanding that, it can be used for any sentence similarity task within the sentence-transformers framework as well. It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search. Take a look at the documentation of sentence-transformers if you want to use this model for more than what we… Open weights 278M parameters 514 tokens sentence-transformers [View model](https://savrn.com/models/lt-un-data-fine-fine-multi) ## Google [All models and datasets](https://savrn.com/model-publishers/google) ## Versions - [57c266a740f5](https://savrn.com/models/embeddinggemma-300m/versions/57c266a740f5) · current 2026-10-23 ## Explore More - [All sentence similarity models](https://savrn.com/models/tasks/sentence-similarity) - [All models under gemma](https://savrn.com/models/licenses/gemma) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-19. - [Google website](https://ai.google.dev/) - [Hugging Face record](https://huggingface.co/google/embeddinggemma-300m) - [How the hub is built](https://savrn.com/model-hub/methodology)