A 23M-Parameter Model Still Wins: 338 Embedding Models Ranked by Real Downloads

all-MiniLM-L6-v2, a 23M-parameter sentence-transformers model, was downloaded 177M times on Hugging Face in the last 30 days. That's 38% of all downloads across the 213 open-weight embedding models tracked by the TextModels directory, and more than the next four models combined. Below are all 338 embedding models in the directory: open weights ranked by downloads, then hosted APIs, Ollama builds and the historical models that started it all. Most "best embedding model" lists rank by benchmark score, like PremAI's ranking by MTEB score, cost and self-hosting. Benchmarks tell you what's best on paper. Downloads tell you what developers actually ship, and the two lists look very different. The sentence-transformers family's lead has a history. As HackerNoon's overview of embeddings for RAG puts it, "Simply said, the BERT model isn't the best for similarity search." SBERT fixed that by embedding each sentence separately so vectors can be computed once and compared fast, and its descendants still dominate the download charts. What the download data says Small and old beats big and new. The top two models by downloads, all-MiniLM-L6-v2 (23M params) and BAAI's bge-small-en-v1.5 (33M), are tiny. For most production search, "fast and good enough" wins. The market is concentrated. The top 10 open-weight embedders account for 74% of the 470M monthly downloads in this list. Qwen is the newcomer to watch. Qwen3-Embedding-0.6B is already #9 with 9M monthly downloads, and its 4B and 8B siblings both clear 2.4M. Multilingual is mainstream. Five of the top 12 are multilingual: bge-m3, two paraphrase-multilingual models and two multilingual-e5 models. Vector databases ship their own builds. Qdrant's ONNX conversions and sparse models appear nine times, which shows how many teams get embeddings through their database instead of picking a model directly. How this list was built Everything comes from the TextModels directory's Embedding filter (338 models) as of November 11, 2026. Hugging Face entries are ranked by 30-day downloads pulled from the Hugging Face API. "Verified" means the TextModels directory fetched and confirmed that model's metrics per model. Hosted API models have no public download counts, so they're grouped by provider, alphabetically. Some hosted entries, like OpenAI's 001-series search models, are long retired. They stay in the list because people still search for them when migrating old pipelines. Part 1: 213 open-weight embedding models on Hugging Face, ranked by 30-day downloads all-MiniLM-L6-v2 by sentence-transformers: 177M downloads/30d, 23M params, Apache 2.0 (Verified) bge-small-en-v1.5 by BAAI: 47.1M downloads/30d, 33M params, MIT (Verified) bge-m3 by BAAI: 29.3M downloads/30d, 568M params, MIT (Verified) paraphrase-multilingual-MiniLM-L12-v2 by sentence-transformers: 26.9M downloads/30d, 118M params, Apache 2.0 (Verified) all-mpnet-base-v2 by sentence-transformers: 20.5M downloads/30d, 109M params, Apache 2.0 (Verified) nomic-embed-text-v1.5 by nomic-ai: 11.9M downloads/30d, 137M params, Apache 2.0 (Verified) multilingual-e5-small by intfloat: 9.4M downloads/30d, 118M params paraphrase-multilingual-mpnet-base-v2 by sentence-transformers: 9.1M downloads/30d, 278M params Qwen3-Embedding-0.6B by Qwen: 9M downloads/30d, 596M params, Apache 2.0 (Verified) bge-base-en-v1.5 by BAAI: 8.4M downloads/30d, 109M params bge-large-en-v1.5 by BAAI: 8.1M downloads/30d, 335M params, MIT (Verified) multilingual-e5-large by intfloat: 7.1M downloads/30d, 560M params, MIT (Verified) granite-embedding-small-english-r2 by ibm-granite: 6.8M downloads/30d, 48M params multilingual-e5-base by intfloat: 4.9M downloads/30d, 278M params all-MiniLM-L6-v2 by Xenova: 4.5M downloads/30d snowflake-arctic-embed-l-v2.0 by Snowflake: 3.8M downloads/30d, 568M params bge-small-zh-v1.5 by BAAI: 3.1M downloads/30d, 24M params bge-base-en-v1.5-course-recommender-v5 by datasocietyco: 2.9M downloads/30d, 109M params gte-small by thenlper: 2.5M downloads/30d, 33M params bge-base-en-v1.5 by Xenova: 2.5M downloads/30d all-MiniLM-L12-v2 by sentence-transformers: 2.5M downloads/30d, 33M params Qwen3-Embedding-4B by Qwen: 2.5M downloads/30d, 4B params all-distilroberta-v1 by sentence-transformers: 2.4M downloads/30d, 82M params Qwen3-Embedding-8B by Qwen: 2.4M downloads/30d, 7.6B params mxbai-embed-large-v1 by mixedbread-ai: 2.3M downloads/30d, 335M params bge-small-en-v1.5-onnx-Q by Qdrant: 2M downloads/30d paraphrase-mpnet-base-v2 by sentence-transformers: 1.9M downloads/30d, 109M params multilingual-e5-large-instruct by intfloat: 1.7M downloads/30d, 560M params all-MiniLM-L6-v2-onnx by Qdrant: 1.6M downloads/30d paraphrase-MiniLM-L6-v2 by sentence-transformers: 1.5M downloads/30d, 23M params multi-qa-mpnet-base-dot-v1 by sentence-transformers: 1.5M downloads/30d, 109M params jina-embeddings-v3 by jinaai: 1.5M downloads/30d, 572M params, CC BY-NC 4.0 (Verified) SapBERT-from-PubMedBERT-fulltext by cambridgeltl: 1.4M downloads/30d, 109M params ko-sroberta-multitask by jhgan: 1.4M downloads/30d, 111M params multilingual-e5-large-onnx by Qdrant: 1.2M downloads/30d jina-embeddings-v2-small-en by jinaai: 1.2M downloads/30d, 33M params embeddinggemma-300m by google: 1.2M downloads/30d, 303M params, Gemma (Verified) e5-small-v2 by intfloat: 1.1M downloads/30d, 33M params e5-base-v2 by intfloat: 1M downloads/30d, 109M params UAE-Large-V1 by WhereIsAI: 1M downloads/30d, 335M params sup-SimCSE-VietNamese-phobert-base by VoVanPhuc: 1M downloads/30d, 136M params LaBSE by sentence-transformers: 986.3K downloads/30d, 471M params vietnamese-bi-encoder by bkai-foundation-models: 983.4K downloads/30d, 135M params e5-base by intfloat: 946.9K downloads/30d, 109M params paraphrase-MiniLM-L3-v2 by sentence-transformers: 945K downloads/30d, 17M params bm25 by Qdrant: 913K downloads/30d gte-multilingual-base by Alibaba-NLP: 891.3K downloads/30d, 305M params nomic-embed-text-v1 by nomic-ai: 882.5K downloads/30d, 137M params distiluse-base-multilingual-cased-v2 by sentence-transformers: 869.3K downloads/30d, 135M params bge-small-en-v1.5 by Xenova: 851.3K downloads/30d text2vec-base-chinese by shibing624: 845.1K downloads/30d, 102M params pubmedbert-base-embeddings by NeuML: 812.2K downloads/30d, 109M params ruri-v3-310m by cl-nagoya: 745.4K downloads/30d, 315M params e5-large-v2 by intfloat: 733.6K downloads/30d, 335M params nomic-embed-text-v2-moe by nomic-ai: 720.4K downloads/30d, 475M params FRIDA by ai-forever: 692.8K downloads/30d, 823M params multi-qa-MiniLM-L6-cos-v1 by sentence-transformers: 659.9K downloads/30d, 23M params specter2_base by allenai: 658.7K downloads/30d embeddinggemma-300m-ONNX by onnx-community: 635.7K downloads/30d all-roberta-large-v1 by sentence-transformers: 610.9K downloads/30d, 355M params nomic-embed-text-v1.5-GGUF by nomic-ai: 574.1K downloads/30d paraphrase-multilingual-MiniLM-L12-v2-onnx-Q by Qdrant: 569.5K downloads/30d plamo-embedding-1b by pfnet: 567.5K downloads/30d, 1.1B params gte-large by thenlper: 555.4K downloads/30d, 335M params snowflake-arctic-embed-xs by Snowflake: 533.9K downloads/30d, 23M params msmarco-bert-base-dot-v5 by sentence-transformers: 526.7K downloads/30d, 109M params langcache-embed-v1 by redis: 525.5K downloads/30d, 149M params bert-base-turkish-cased-mean-nli-stsb-tr by emrecan: 520.6K downloads/30d, 111M params deepset-mxbai-embed-de-large-v1 by mixedbread-ai: 514.5K downloads/30d, 487M params snowflake-arctic-embed-l by Snowflake: 506.6K downloads/30d, 334M params bge-small-en-v1.5 by unsloth: 469.6K downloads/30d, 33M params rubert-tiny2 by cointegrated: 467.7K downloads/30d, 29M params multilingual-e5-small by Xenova: 457.8K downloads/30d gte-base-en-v1.5 by Alibaba-NLP: 452.6K downloads/30d, 137M params nli-mpnet-base-v2 by sentence-transformers: 452K downloads/30d, 109M params llama2-embedding-1b-8k by mesolitica: 443.8K downloads/30d, 947M params S-PubMedBert-MedQuAD by TimKond: 436.7K downloads/30d, 109M params efficient-splade-VI-BT-large-doc by naver: 427K downloads/30d LLM2Vec-Meta-Llama-3-8B-Instruct-mntp by McGill-NLP: 405.6K downloads/30d, 8B params BioLORD-2023 by FremyCompany: 397.1K downloads/30d, 109M params snowflake-arctic-embed-m by Snowflake: 387.3K downloads/30d, 109M params jina-embeddings-v2-base-en by jinaai: 356.8K downloads/30d, 137M params llama-nemotron-embed-1b-v2 by nvidia: 347.3K downloads/30d, 1.2B params paraphrase-albert-small-v2 by sentence-transformers: 340K downloads/30d, 12M params nomic-embed-code by nomic-ai: 336K downloads/30d, 7.1B params bge-large-zh-v1.5 by BAAI: 334.9K downloads/30d Nemotron-3-Embed-1B-BF16 by nvidia: 324.8K downloads/30d, 1.1B params paraphrase-multilingual-MiniLM-L12-v2 by Xenova: 314.8K downloads/30d Solon-embeddings-large-0.1 by OrdalieTech: 311.4K downloads/30d, 560M params bert-base-nli-mean-tokens by sentence-transformers: 308.2K downloads/30d, 109M params USER-bge-m3 by deepvk: 288.6K downloads/30d, 359M params multi-qa-MiniLM-L6-dot-v1 by sentence-transformers: 286.2K downloads/30d, 23M params msmarco-MiniLM-L6-v3 by sentence-transformers: 284.3K downloads/30d, 23M params e5-large by intfloat: 263.5K downloads/30d, 335M params all_miniLM_L6_v2_with_attentions by Qdrant: 250.4K downloads/30d indobert-base-p1 by indobenchmark: 247.8K downloads/30d KURE-v1 by nlpai-lab: 245.5K downloads/30d, 568M params mmlw-retrieval-roberta-large by sdadas: 236K downloads/30d, 435M params sentence-transformers-alephbert by imvladikon: 235.4K downloads/30d dragon-multiturn-query-encoder by nvidia: 231K downloads/30d dragon-multiturn-context-encoder by nvidia: 230.8K downloads/30d codebert-base by microsoft: 227.3K downloads/30d msmarco-distilbert-base-tas-b by sentence-transformers: 226.2K downloads/30d, 66M params harrier-oss-v1-0.6b by microsoft: 222K downloads/30d, 596M params bge-small-en by BAAI: 219.4K downloads/30d, 33M params mxbai-embed-large-v1-gguf by ChristianAzinn: 215.3K downloads/30d bart-base by facebook: 213.8K downloads/30d, 139M params sbert-legal-xlm-roberta-base by Stern5497: 205.6K downloads/30d snowflake-arctic-embed-l-v2.0-ko by dragonkue: 205.5K downloads/30d, 568M params granite-embedding-125m-english by ibm-granite: 205.1K downloads/30d, 125M params SapBERT-UMLS-2020AB-all-lang-from-XLMR by cambridgeltl: 202.1K downloads/30d, 278M params LLM2Vec-Meta-Llama-3-8B-Instruct-mntp-supervised by McGill-NLP: 196.5K downloads/30d, 8B params instructor-large by hkunlp: 189.7K downloads/30d paraphrase-xlm-r-multilingual-v1 by sentence-transformers: 181.4K downloads/30d, 278M params German_Semantic_STS_V2 by aari1995: 180.9K downloads/30d, 336M params paraphrase-MiniLM-L12-v2 by sentence-transformers: 174.9K downloads/30d, 33M params msmarco-MiniLM-L12-cos-v5 by sentence-transformers: 174.5K downloads/30d, 33M params All-MiniLM-L6-v2-Embedding-GGUF by second-state: 169.3K downloads/30d unixcoder-base by microsoft: 167.4K downloads/30d all-MiniLM-L6-v2-GGUF by leliuga: 165.5K downloads/30d distilbert-base-nli-stsb-mean-tokens by sentence-transformers: 156.4K downloads/30d, 66M params multi-qa-mpnet-base-cos-v1 by sentence-transformers: 153.1K downloads/30d, 109M params bge-base-en-v1.5-onnx-Q by Qdrant: 152.7K downloads/30d paraphrase-multilingual-minilm-l12-v3-mn by gmunkhtur: 151.7K downloads/30d, 118M params gte-base by thenlper: 151K downloads/30d, 109M params jina-embeddings-v2-base-code by jinaai: 146.5K downloads/30d, 161M params BGE-m3-ko by dragonkue: 143K downloads/30d, 568M params ModernBERT-base-msmarco by joe32140: 141K downloads/30d, 149M params Nemotron-3-Embed-8B-BF16 by nvidia: 138.9K downloads/30d, 8B params S-PubMedBert-MS-MARCO by pritamdeka: 138.5K downloads/30d canine-c by google: 136.6K downloads/30d, 132M params sentence-t5-base by sentence-transformers: 131.2K downloads/30d, 110M params gte-modernbert-base by Alibaba-NLP: 127.3K downloads/30d, 149M params ruri-v3-30m by cl-nagoya: 125.6K downloads/30d, 37M params mxbai-embed-2d-large-v1 by mixedbread-ai: 120.3K downloads/30d, 335M params snowflake-arctic-embed-m-v1.5 by Snowflake: 116.5K downloads/30d, 109M params modernbert-embed-base by nomic-ai: 115.1K downloads/30d, 149M params all-indo-e5-small-v4 by LazarusNLP: 114.2K downloads/30d, 118M params gte-Qwen2-1.5B-instruct by Alibaba-NLP: 111.5K downloads/30d, 1.8B params harrier-oss-v1-270m by microsoft: 105.2K downloads/30d, 268M params GIST-Embedding-v0 by avsolatorio: 104.5K downloads/30d, 109M params bart-large by facebook: 103.5K downloads/30d herbert-base-cased by allegro: 103.4K downloads/30d e5-base-multilingual-4096 by efederici: 98.5K downloads/30d voyage-4-nano by voyageai: 93.4K downloads/30d, 346M params ko-sbert-nli by jhgan: 91.4K downloads/30d GIST-all-MiniLM-L6-v2 by avsolatorio: 90.9K downloads/30d, 23M params jina-embeddings-v5-text-small-retrieval by jinaai: 89.1K downloads/30d, 596M params bge-m3-spa-law-qa by littlejohn-ai: 87.8K downloads/30d, 568M params vietnamese-document-embedding by dangvantuan: 86.5K downloads/30d, 305M params granite-embedding-311m-multilingual-r2 by ibm-granite: 86.5K downloads/30d, 312M params ko-sbert-sts by jhgan: 83.5K downloads/30d jina-embeddings-v5-text-nano by jinaai: 83.1K downloads/30d, 212M params ruri-v3-130m by cl-nagoya: 81K downloads/30d, 132M params e5-mistral-7b-instruct by intfloat: 79.8K downloads/30d, 7.1B params jina-embeddings-v5-text-small by jinaai: 79.8K downloads/30d, 596M params msmarco-distilbert-base-v4 by sentence-transformers: 79K downloads/30d, 66M params Nemotron-3-Embed-1B-Q4_K_M-GGUF by zenmagnets: 78.3K downloads/30d, 1B params rubert-base-cased by DeepPavlov: 77.2K downloads/30d KR-SBERT-V40K-klueNLI-augSTS by snunlp: 76.5K downloads/30d bge-m3-GGUF by gpustack: 71.8K downloads/30d GIST-small-Embedding-v0 by avsolatorio: 70.3K downloads/30d, 33M params st-codesearch-distilroberta-base by flax-sentence-embeddings: 67.6K downloads/30d Splade_PP_en_v1 by Qdrant: 65.8K downloads/30d multi-qa-distilbert-dot-v1 by sentence-transformers: 63.8K downloads/30d, 66M params stsb-xlm-r-multilingual by sentence-transformers: 63.5K downloads/30d, 278M params embeddinggemma-300m by unsloth: 62K downloads/30d, 303M params multi-qa-distilbert-cos-v1 by sentence-transformers: 61K downloads/30d, 66M params lilt-roberta-en-base by SCUT-DLVCLab: 60.4K downloads/30d, 131M params bge-m3 by Xenova: 56.5K downloads/30d stella_en_400M_v5 by NovaSearch: 56.1K downloads/30d, 435M params bge-multilingual-gemma2 by BAAI: 53.1K downloads/30d, 9.2B params snowflake-arctic-embed-m-v2.0 by Snowflake: 50K downloads/30d, 305M params sentence-transformer-klue by hunkim: 47.9K downloads/30d granite-embedding-30m-english by ibm-granite: 46.8K downloads/30d, 30M params granite-embedding-107m-multilingual by ibm-granite: 45.4K downloads/30d, 107M params snowflake-arctic-embed-s by Snowflake: 43.6K downloads/30d, 33M params text2vec-base-chinese-paraphrase by shibing624: 42.3K downloads/30d, 118M params granite-embedding-278m-multilingual by ibm-granite: 41.6K downloads/30d, 278M params sentence-bert-swedish-cased by KBLab: 41.3K downloads/30d, 125M params colbert-bert-tiny by NeuML: 36.4K downloads/30d, 4M params jina-embeddings-v5-text-nano-retrieval by jinaai: 35.6K downloads/30d, 212M params indo-sentence-bert-base by firqaaa: 32.8K downloads/30d paraphrase-distilroberta-base-v1 by sentence-transformers: 32.1K downloads/30d, 82M params bge-large-en-v1.5-onnx by Qdrant: 31.4K downloads/30d USER-base by deepvk: 29.8K downloads/30d, 124M params gte-Qwen2-7B-instruct by Alibaba-NLP: 26.9K downloads/30d, 7.6B params text2vec-base-multilingual by shibing624: 25.3K downloads/30d, 118M params Giga-Embeddings-instruct-480M-0826 by ai-sage: 22.4K downloads/30d, 484M params st-polish-paraphrase-from-mpnet by sdadas: 13.8K downloads/30d, 124M params mdbr-leaf-ir by MongoDB: 13.8K downloads/30d, 23M params vietnamese-sbert by keepitreal: 11.2K downloads/30d llama-nemotron-embed-vl-1b-v2 by nvidia: 9.4K downloads/30d, 1.7B params stsb-roberta-base-v2 by sentence-transformers: 7.4K downloads/30d, 125M params all-MiniLM-L6-v2-4bit by mlx-community: 7.3K downloads/30d, 23M params LaBSE by setu4993: 6.8K downloads/30d, 471M params GLuCoSE-base-ja-v2 by pkshatech: 6K downloads/30d, 133M params LaBSE-en-ru by cointegrated: 5.6K downloads/30d, 129M params Octen-Embedding-8B by Octen: 4.6K downloads/30d, 7.6B params jina-embeddings-v5-text-small-text-matching by jinaai: 4.2K downloads/30d, 596M params PubMedBERT-mnli-snli-scinli-scitail-mednli-stsb by pritamdeka: 3.8K downloads/30d Giga-Embeddings-instruct by ai-sage: 3.6K downloads/30d, 3.5B params Qwen3-Embedding-4B-W4A16-G128 by boboliu: 2.1K downloads/30d, 4B params serafim-900m-portuguese-pt-sentence-encoder-ir by PORTULAN: 1.4K downloads/30d, 885M params Solon-embeddings-base-0.1 by OrdalieTech: 685 downloads/30d, 278M params LEALLA-base by setu4993: 569 downloads/30d, 107M params sarashina-embedding-v1-1b by sbintuitions: 417 downloads/30d, 1.2B params minilm-bo-en-sim by khyentsevision: 407 downloads/30d, 23M params stella_en_400M_v5 by billatsectorflow: 240 downloads/30d, 435M params OTel-Embedding-109M by farbodtavakkoli: 179 downloads/30d, 109M params OTel-Embedding-34M by farbodtavakkoli: 123 downloads/30d, 33M params OTel-Embedding-300M by farbodtavakkoli: 53 downloads/30d, 303M params OTel-Embedding-33M by farbodtavakkoli: 51 downloads/30d, 33M params Part 2: 55 hosted embedding APIs, by provider These are models you call over an API instead of downloading. The list covers current models and retired ones still found in old codebases. OpenAI (19) 214. text-embedding-3-large 215. text-embedding-3-small 216. text-embedding-ada-002 217. code-search-ada-code-001 218. code-search-ada-text-001 219. code-search-babbage-code-001 220. code-search-babbage-text-001 221. text-search-ada-doc-001 222. text-search-ada-query-001 223. text-search-babbage-doc-001 224. text-search-babbage-query-001 225. text-search-curie-doc-001 226. text-search-curie-query-001 227. text-search-davinci-doc-001 228. text-search-davinci-query-001 229. text-similarity-ada-001 230. text-similarity-babbage-001 231. text-similarity-curie-001 232. text-similarity-davinci-001 Cohere (8) 233. embed-v4.0 234. embed-english-v3.0 235. embed-english-light-v3.0 236. embed-multilingual-v3.0 237. embed-multilingual-light-v3.0 238. embed-english-v2.0 239. embed-english-light-v2.0 240. embed-multilingual-v2.0 Google Vertex AI (8) 241. gemini-embedding-001 242. gemini-embedding-002 243. text-embedding-004 244. text-multilingual-embedding-002 245. textembedding-gecko@003 246. textembedding-gecko@002 247. textembedding-gecko@001 248. textembedding-gecko-multilingual@001 NVIDIA NIM (7) 249. nv-embed-v2 250. nv-embed-v1 251. llama-3.2-nv-embedqa-1b-v2 252. llama-3.2-nv-embedqa-1b-v1 253. nv-embedqa-mistral-7b-v2 254. nv-embedqa-e5-v5 255. nv-embedcode-7b-v1 IBM watsonx / Granite (6) 256. granite-embedding-english-r2 257. granite-embedding-multilingual-r2 258. granite-embedding-30m-english 259. granite-embedding-125m-english 260. granite-embedding-107m-multilingual 261. granite-embedding-278m-multilingual Amazon Bedrock (4) 262. amazon.titan-embed-text-v2:0 263. amazon.titan-embed-text-v2:1 264. amazon.titan-embed-text-v1 265. amazon.titan-embed-g1-text-02 Mistral AI (3) 266. mistral-embed 267. codestral-embed-2506 268. codestral-embed-2505 Part 3: 20 embedding models you can run locally with Ollama Ollama doesn't publish download counts, so these are sorted from smallest to largest. Pick the smallest one that clears your quality bar. 269. all-minilm 22m 270. snowflake-arctic-embed 22m 271. granite-embedding 30m (IBM) 272. all-minilm 33m 273. snowflake-arctic-embed 33m 274. snowflake-arctic-embed 110m 275. nomic-embed-text 137m 276. snowflake-arctic-embed 137m 277. granite-embedding 278m (IBM) 278. paraphrase-multilingual 278m 279. embeddinggemma 300m (Google) 280. bge-large 335m 281. mxbai-embed-large 335m 282. snowflake-arctic-embed 335m 283. bge-m3 567m 284. snowflake-arctic-embed2 568m 285. qwen3-embedding 0.6b (Alibaba / Qwen) 286. qwen3-embedding 4b (Alibaba / Qwen) 287. qwen3-embedding 8b (Alibaba / Qwen) 288. nomic-embed-text-v2-moe latest (size not listed) Part 4: 50 historical embedding models that got us here Before sentence-transformers there were word vectors, and before RAG there was semantic search. These are the ancestors of everything above. 289. word2vec Skip-gram Google News (Google) 290. word2vec CBOW Google News (Google) 291. GloVe 840B 300d (Stanford NLP) 292. GloVe 42B 300d (Stanford NLP) 293. GloVe 6B 300d (Stanford NLP) 294. GloVe 6B 200d (Stanford NLP) 295. GloVe 6B 100d (Stanford NLP) 296. GloVe 6B 50d (Stanford NLP) 297. GloVe Twitter 27B 200d (Stanford NLP) 298. GloVe Twitter 27B 100d (Stanford NLP) 299. GloVe Twitter 27B 50d (Stanford NLP) 300. GloVe Twitter 27B 25d (Stanford NLP) 301. fastText Common Crawl 300d 2M (Meta AI) 302. fastText Wiki News 300d 1M (Meta AI) 303. ELMo Original 5.5B (Allen Institute for AI) 304. InferSent v1 (Meta AI) 305. Universal Sentence Encoder DAN (Google) 306. Universal Sentence Encoder Transformer (Google) 307. LASER2 (Meta AI) 308. bert-base-uncased (Google) 309. bert-base-cased (Google) 310. bert-large-uncased (Google) 311. bert-large-cased (Google) 312. albert-base-v1 (Google) 313. albert-base-v2 (Google) 314. albert-large-v1 (Google) 315. albert-large-v2 (Google) 316. albert-xlarge-v1 (Google) 317. albert-xlarge-v2 (Google) 318. albert-xxlarge-v1 (Google) 319. albert-xxlarge-v2 (Google) 320. Sentence-BERT Base NLI Mean Tokens (UKP Lab) 321. Sentence-RoBERTa Large NLI STS-B (UKP Lab) 322. SimCSE BERT Base Supervised (Princeton) 323. SimCSE BERT Base Unsupervised (Princeton) 324. SimCSE RoBERTa Large Supervised (Princeton) 325. SimCSE RoBERTa Base Unsupervised (Princeton) 326. SapBERT (Cambridge) 327. RetroMAE Base (Microsoft) 328. GTR Base (Google) 329. GTR Large (Google) 330. GTR XL (Google) 331. GTR XXL (Google) 332. SGPT 125M (Muennighoff) 333. SGPT 1.3B (Muennighoff) 334. SGPT 2.7B (Muennighoff) 335. SGPT 5.8B (Muennighoff) 336. INSTRUCTOR Base (HKU NLP) 337. INSTRUCTOR XL (HKU NLP) 338. SONAR Text Encoder (Meta AI) So which one should you use? If you need a default today, the download data points to all-MiniLM-L6-v2 or bge-small-en-v1.5 for English, bge-m3 or multilingual-e5 for multilingual, and Qwen3-Embedding for top quality if you have the GPU. Then test on your own data. The model is only half the stack; HackerNoon's guide to choosing a vector database for a production RAG chatbot covers the other half. To compare all 3,009 text models, including rerankers, classifiers and the LLMs these embeddings feed, browse the TextModels directory.