mxbai-embed-large-v1
Free.ai (self-hosted)
·
embeddings
·
~100 tokens per call
mxbai-embed-large-v1 is an embedding model built by mixedbread.ai. Strongest at Semantic search, clustering, similarity.. Self-inotarisirwa pa Free.ai GPUs - inofamba zvakasununguka pazuva rako re token pool (100 tokens per call). Yakaburitswa pasi peApache 2.0 - kushandiswa kwekutengesa kwakabvumirwa pasi peFree.ai.
Zvimwe zvinobvunzwa kakawanda
mxbai-embed-large-v1 inoshandura mazita ezvinyorwa kuita mazita ezvinyorwa (mazita ezvinyorwa) ayo anotora pfungwa. Usaishandisa kune semantic kutsvaka, kuunganidza, kukurudzira, kudzokorora-kuwedzera kuumbwa (RAG), uye chero basa uko "izvi zvinyorwa zvakafanana nezvinyorwa" zvinokosha.
Zviri pachena kuti mavara anowanzova 384, 768, 1024, kana 1536 zvichienderana nechigadzirwa. BGE-M3 inoburitsa 1024-dim; OpenAI Ada inoburitsa 1536. The API response inosanganisira mavara kuti yako vector DB inotora yakakodzera index.
Modern embedding models (kusanganisira zvakawanda sarudzo pa Free.ai) vanodzidziswa pa 100 + zvinyorwa. Cross-chirungu kudzoka mabasa - tsvaga muChirungu, mutambo mapepa muSpanish.
512 kusvika 8,192 tokens zvichienderana nemhando. Yakareba zvinyorwa zvinovharwa — chunk refu mapepa mumaparagiramu pamberi embedding.
mxbai-embed-large-v1 inoita basa paGPU yedu uye inosanganisira maturusi anodhura - pamusoro pe100 tokens pafoni yakagadzirwa kubva kune yako yezuva nezuva yemahara pool. $5 = 200K tokens.
Yeah - POST a list of strings to /v1/embeddings/ and mxbai-embed-large-v1 returns a list of vectors in the same order. Batch size up to 2,048 per request.
L2-normalized by default — cosine similarity = dot product. Pass `normalize=false` kana iwe uchida raw vectors yeimwe distance metric.
Any - Pinecone, Weaviate, Qdrant, Chroma, pgvector, FAISS, LanceDB. mxbai-embed-large-v1 anodzokera plain JSON floats; DB kamwe haasi kuona chifananidzo.
Yeah - POST to /v1/embeddings/ with model="mxbai-embed-large-v1". OpenAI-compatible response shape, so existing client libraries work unchanged. /api/ has the full reference.
Self-hosted mamodheru kuchengetedza yako tenzi paGPUs yedu uye kubvisa mushure mekudzoka kufona. Premium kubuda ne DPA. Hatina kudzidzisa pazvinhu zvenyu.
Sub-100ms yenyaya pfupi pa self-hosted, 100-500ms papremium. Batch kufona kukwira zvakapetwa kaviri - 1,000 mabhureki akazara mu 2-10 masekondi.
Yeah - Free.ai grants commercial use of embeddings. Build production search, RAG pipelines, recommendation systems with no per-vector royalty.