{"v":1,"id":"model:hf:ngquocvinh/EmbeddingGemma-2-GGUF","slug":"model-ngquocvinh-embeddinggemma-2-gguf","kind":"model","category":"embedding","title":"EmbeddingGemma-2-GGUF","summary":"Community GGUF quantizations of google/embeddinggemma-2.","source":{"provider":"hf","ref":"ngquocvinh/EmbeddingGemma-2-GGUF","url":"https://huggingface.co/ngquocvinh/EmbeddingGemma-2-GGUF","rev":"8e83fab7c71a90d8f97ae7cacf6a488f654cb5b6","fetchedAt":"2026-10-09T19:01:02.077Z","etag":"W/\"25a5-Z+3bkAmwx5ud9t+zPxEx90QD5fs\""},"author":{"name":"ngquocvinh","url":"https://huggingface.co/ngquocvinh"},"license":{"spdx":"apache-2.0","raw":"apache-2.0","open":true},"metrics":{"downloads":4365,"downloadsWeek":297971,"likes":1,"stars":5776,"openIssues":330,"lastRelease":{"tag":"v4.0.1","at":"2026-05-20T15:37:55Z"},"pushedAt":"2026-10-06T09:16:03Z","takenAt":"2026-10-09T19:01:02.077Z"},"tags":["llama.cpp","gguf","embedding","multimodal","multilingual","quantized","sentence-transformers","feature-extraction","endpoints_compatible","imatrix","conversational"],"pipeline":"feature-extraction","links":{"github":"google-deepmind/gemma","npm":"node-llama-cpp"},"updatedAt":"2026-10-08T04:18:29.000Z","collectedAt":"2026-10-09T19:01:02.077Z","review":{"numbers":["4,365 downloads on Hugging Face","1 likes","license apache-2.0","0.2 GB for EmbeddingGemma-2-Q4_K_M.gguf","5,776 stars on google-deepmind/gemma","330 open issues and PRs","last release v4.0.1 on 2026-05-20","297,971 npm downloads a week for node-llama-cpp","latest node-llama-cpp@3.22.1"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-09T19:01:02.077Z","http":200},"description":"# EmbeddingGemma 2 GGUF\n\nCommunity GGUF quantizations of [google/embeddinggemma-2](https://huggingface.co/google/embeddinggemma-2).\n\n☕ If this GGUF made your day easier, a coffee would make mine.\nSend a coffee ☕\nI build and test these releases myself. Your coffee helps keep me going.\nThank you for supporting this work.\n\n## About EmbeddingGemma 2\n\n[EmbeddingGemma 2](https://huggingface.co/google/embeddinggemma-2) is a multilingual, multimodal embedding model from Google DeepMind. It maps text and code, images, video, and audio into a shared 768-dimensional vector space and has an 8,192-token shared context window. The upstream checkpoint has a 270M-parameter text path plus optional vision and audio encoders. This release contains a quantized text backbone and a shared BF16 vision/audio projector. Text, image, and audio inputs returned normalized 768-dimensional vectors in local `llama-server` CPU smoke checks. See the [official model card](https://huggingface.co/google/embeddinggemma-2) for supported task prefixes, modalities, and input guidance.\n\n## Embedding evaluation\n\nEvery value below comes from local measurements of the locked BF16 GGUF and quantized files; the numbers are not copied from the upstream model card. The fixed task evaluation uses the 1,379-pair test split of [MTEB STSBenchmark STS](https://huggingface.co/datasets/mteb/stsbenchmark-sts), dataset revision `96943a16ea6a35129e253c659081cb59daf81b30`. It covers 2,552 unique sentences with the `task: sentence similarity | query:` prefix, an 8,192-token context, and the CPU `llama.cpp` runtime at commit `9c2e0e491a822adae1f0b1c831adb4160057d24f`. Spearman and Pearson measure correlation between cosine similarity and human scores. Mean and 5th-percentile cosine measure vector agreement with the BF16 GGUF reference. Higher values indicate stronger task correlation or closer vector agreement; smaller files use less disk. This is a task-specif…\n\nSource: https://huggingface.co/ngquocvinh/EmbeddingGemma-2-GGUF","install":{"kind":"model","hfId":"ngquocvinh/EmbeddingGemma-2-GGUF","gated":false,"format":"gguf","files":[{"name":"EmbeddingGemma-2-IQ1_M.gguf","size":103040640,"quant":"IQ1_M","sha256":"a7788cc0daf653026d9762fc90e351763e472427cb2ac9eef1cdb3d89538057a"},{"name":"EmbeddingGemma-2-IQ1_S.gguf","size":100558464,"quant":"IQ1_S","sha256":"df675902e617d89e13b87412d5cea3f562cb3bcad4d7a3942726f44a58ac3d49"},{"name":"EmbeddingGemma-2-IQ2_M.gguf","size":130909824,"quant":"IQ2_M","sha256":"104366160074408cdfef0d8322aeddf4453ab5556f768bd47387aa316c92e3f5"},{"name":"EmbeddingGemma-2-IQ2_S.gguf","size":127600256,"quant":"IQ2_S","sha256":"c4af164cb11d7380e5d905084c08daed021d58407a4004c9c230cc2929bdd2d8"},{"name":"EmbeddingGemma-2-IQ2_XS.gguf","size":110945920,"quant":"IQ2_XS","sha256":"c4e59a215a94e4ab077e1a10c43df85ab1c63addc8fdf5515a5ee9b99c8a15f4"},{"name":"EmbeddingGemma-2-IQ2_XXS.gguf","size":107177600,"quant":"IQ2_XXS","sha256":"b2ffd17ae3e7c1e54048a7f6c073848eb05d6b5554e7e1f89c52e607c811e79a"},{"name":"EmbeddingGemma-2-IQ3_M.gguf","size":145748608,"quant":"IQ3_M","sha256":"173b0e1daabbd79873c5757dc0d44cefb269927286b978f4685f5232831ac90a"},{"name":"EmbeddingGemma-2-IQ3_S.gguf","size":142545536,"quant":"IQ3_S","sha256":"ce15d1c47c0b9bd3ab51921b8fbf4758b767fe8ae035a81b44783e37d01a652d"},{"name":"EmbeddingGemma-2-IQ3_XS.gguf","size":139793024,"quant":"IQ3_XS","sha256":"ba5ce4f631e5a9515c6dcd41b2f96d2ec74ef829d1dd0a4e541efc633ff8c387"},{"name":"EmbeddingGemma-2-IQ3_XXS.gguf","size":135775872,"quant":"IQ3_XXS","sha256":"5d743354ee1458ded40169cf549f93b7ec71e58f3ed6b909abcdda2c27664149"},{"name":"EmbeddingGemma-2-IQ4_NL.gguf","size":177640064,"quant":"IQ4_NL","sha256":"1ae7f62c21191734a7c8c4f2708537e29e5ca99bfe354284551ba8249f975e80"},{"name":"EmbeddingGemma-2-IQ4_XS.gguf","size":169382528,"quant":"IQ4_XS","sha256":"96db13a1041a7ef09bf0ca3a7e582b3331dae326a3aca9898a93f257979af427"},{"name":"EmbeddingGemma-2-Q1_0.gguf","size":66163328,"quant":"Q1_0","sha256":"4953a59ae0d8a725052407f458101ef81d8acb6a95f3293f4949ff3f594ef18a"},{"name":"EmbeddingGemma-2-Q2_K.gguf","size":120394368,"quant":"Q2_K","sha256":"67124cbba678cce4664bf9af72839a88ab12a9083821545c56847f41a54e96a1"},{"name":"EmbeddingGemma-2-Q2_K_S.gguf","size":116445824,"quant":"Q2_K_S","sha256":"35cb1027140ea72970fc227553464cb901d581d658670f707ed4637564e1de48"},{"name":"EmbeddingGemma-2-Q3_K_L.gguf","size":154440320,"quant":"Q3_K_L","sha256":"a32206e9431dfc310a4eab442ef98bb87be02c9eef03efb4ad6b6dd2b553e4da"},{"name":"EmbeddingGemma-2-Q3_K_M.gguf","size":148869760,"quant":"Q3_K_M","sha256":"a5c902b15458c4886d6f1296bad71fd8084d7ac626cb8713778cf2af5e1b5a71"},{"name":"EmbeddingGemma-2-Q3_K_S.gguf","size":142545536,"quant":"Q3_K_S","sha256":"a6c72c2febc19a36afa8de82fb7e6cac8a45a2e7a20922612b2cc7c8b19270f6"},{"name":"EmbeddingGemma-2-Q4_K_M.gguf","size":181695104,"quant":"Q4_K_M","sha256":"ea25550b050fa8826707c5e44659a7c18259606dcc6ba449f771e3c75cf4e663"},{"name":"EmbeddingGemma-2-Q4_K_S.gguf","size":178164352,"quant":"Q4_K_S","sha256":"4a9b92529b930be9f6cc858821b5d6ceea46c9631a05bd27f33a6942f46b63f6"},{"name":"EmbeddingGemma-2-Q5_K_M.gguf","size":212759168,"quant":"Q5_K_M","sha256":"22f79226a02bfa2dd840414f342a004f93b3cb85e498923a5e089f03f6410b1f"},{"name":"EmbeddingGemma-2-Q5_K_S.gguf","size":210670208,"quant":"Q5_K_S","sha256":"0c0dc8e88bc69c720f997ea1dd9cf4a57bbbccaecfa85925455a6608654d2d81"},{"name":"EmbeddingGemma-2-Q6_K.gguf","size":245764736,"quant":"Q6_K","sha256":"21693ca41877ffb20727763e167c503b0467eac4ccef51551230d72c2afcd37e"},{"name":"EmbeddingGemma-2-Q8_0.gguf","size":309855872,"quant":"Q8_0","sha256":"eec0ccc2e76aa85b415651e869c25e476279d6b7d1985533c8296f13989efa10"},{"name":"EmbeddingGemma-2-TQ1_0.gguf","size":131928704,"quant":"TQ1_0","sha256":"a55f16a919b1e31c9bd15ca957167b57437d9d91efc6bd5badb8980e294f6a1c"},{"name":"EmbeddingGemma-2-TQ2_0.gguf","size":138023552,"quant":"TQ2_0","sha256":"539fc9cd918d02b91a4cdb5953436df380c0bc74fd945ef85808935f47482542"},{"name":"mmproj-EmbeddingGemma-2-BF16.gguf","size":982074880,"quant":"BF16","sha256":"b663b6822d1f2d641acc54fc0adfc143b630b7ac87ec54fd60888f325071a0de"}],"totalBytes":4930914048,"suggestedFile":"EmbeddingGemma-2-Q4_K_M.gguf","requirements":{"ramGb":1,"diskBytes":181695104,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["llama.cpp"],"command":"lsh models install hf:ngquocvinh/EmbeddingGemma-2-GGUF"}}