{"work":{"id":"e83995d5-78dc-4a7b-8d9a-1776d004351d","openalex_id":"https://openalex.org/W4415330695","doi":"10.48550/arxiv.2509.20354","arxiv_id":"2509.20354","raw_key":null,"title":"EmbeddingGemma: Powerful and Lightweight Text Representations","authors":null,"authors_text":"Henrique Schechter Vera, Sahil Dua, Biao Zhang, Daniel Salz, Ryan Mullins, Sindhu Raghuram Panyam","year":2025,"venue":"cs.CL","abstract":"We introduce EmbeddingGemma, a new lightweight, open text embedding model based on the Gemma 3 language model family. Our innovative training recipe strategically captures knowledge from larger models via encoder-decoder initialization and geometric embedding distillation. We improve model robustness and expressiveness with a spread-out regularizer, and ensure generalizability by merging checkpoints from varied, optimized mixtures. Evaluated on the Massive Text Embedding Benchmark (MTEB) across multilingual, English, and code domains, EmbeddingGemma (300M) achieves state-of-the-art results. Notably, it outperforms prior top models, both proprietary and open, with fewer than 500M parameters, and provides performance comparable to models double its size, offering an exceptional performance-to-cost ratio. Remarkably, this lead persists when quantizing model weights or truncating embedding outputs. This makes EmbeddingGemma particularly well-suited for low-latency and high-throughput use cases such as on-device applications. We provide ablation studies exploring our key design choices. We release EmbeddingGemma to the community to promote further research.","external_url":"https://arxiv.org/abs/2509.20354","cited_by_count":0,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2509.20354","created_at":"2026-05-08T17:28:41.967130+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"EmbeddingGemma: Powerful and Lightweight Text Representations","render_title":"EmbeddingGemma: Powerful and Lightweight Text Representations"},"hub":{"state":{"work_id":"e83995d5-78dc-4a7b-8d9a-1776d004351d","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":56,"external_cited_by_count":0,"distinct_field_count":11,"first_pith_cited_at":"2025-10-08T14:16:20+00:00","last_pith_cited_at":"2026-07-09T16:50:54+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-22T05:09:27.242525+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":4},{"context_role":"baseline","n":2},{"context_role":"method","n":2}],"polarity_counts":[{"context_polarity":"background","n":3},{"context_polarity":"baseline","n":2},{"context_polarity":"use_method","n":2},{"context_polarity":"unclear","n":1}],"runs":{},"summary":{},"graph":{},"authors":[]}}