Nomic Embed v2
Free.ai (self-hosted)
·
embeddings
·
~100 Ii-token call
Nomic Embed v2 yi an I-embedding model eyenziwe ngu Nomic AI. Enamandla kakhulu kwi Retrieval augmented generation with flexible vector sizes.. I-self-hosted kwi-Free.ai GPUs — iqhuba ngokukhululekileyo kwi-token pool yakho yosuku (100 tokens Iinkcukacha). Ikhutshwe phantsi kwe Apache 2.0 - ukusetyenziswa korhwebo kuvunyelwe kwi Free.ai.
Imibuzo ebuzwa rhoqo
Nomic Embed v2 iguqula umbhalo ube ngumgca ohamba ngokuthe nkqo (uluhlu lwee-floats) othatha uphawu. Sebenzisa ukukhangela okunophawu, ukudibanisa, ukucebisa, ukuphinda ubuyisele- uphuhliso olukhulu (RAG), nanoma yiphi imisebenzi apho "ukuba lo mbhalo ufana nombhalo" ubalulekile.
Iinkcukacha eziqhelekileyo ziyi 384, 768, 1024, okanye 1536 kuxhomekeke kwimodeli. BGE-M3 ikhupha 1024-dim; OpenAI Ada ikhupha 1536. Uphendulo lwe API luquka inkcukacha ukuze i-vector DB yakho ikhethe isalathisi esifanelekileyo.
Iimodeli zokufaka ezitsha (eziquka ezininzi iinketho kwi Free.ai) ziqeqeshwe kwiilwimi ezingaphezu kwe100. Ukufumana kwakhona iilwimi ezingqukuva kusebenza - khangela ngesiNgesi, dibanisa amaxwebhu ngesiSpanish.
512 to 8,192 tokens depending on the model. Longer inputs are truncated - chunk long documents into paragraphs before embedding.
Nomic Embed v2 iqhuba kwiGPUs zethu zazo naphakathi kwezixhobo ezibizayo - malunga ~100 iimpawu zoqhagamshelwano olusuka kwi-pool yakho yosuku ngalunye. $5 = 200K iimpawu.
Ewe - UTHENGA uluhlu lwamagama kwi /v1/embeddings/ kwaye Nomic Embed v2 ibuyisela uluhlu lwee-vectors ngokulandelana okufanayo. Ubungakanani beqela ukuya kuthi ga kwi-2,048 ngesicelo ngasinye.
L2-iqhelekileyo ngokungagqibekanga - uthelekiso lwe-cosinus = ukwenziwa kwe-dot. Yiya `iqhelekileyo=ubuxoki` ukuba ufuna i-raw vectors ye-different distance metric.
Nayiphi na — Pinecone, Weaviate, Qdrant, Chroma, pgvector, FAISS, LanceDB. Nomic Embed v2 ibuyisela i-JSON eqhelekileyo ejikelezayo; i-DB ayizukubona imodeli.
Ewe - UTHANDO ku /v1/ embeddings/ ngemodeli "Nomic Embed v2". Uhlobo lwempendulo oluhambelana ne OpenAI, ngoko iilayibrari zomxhasi ezikhoyo zisebenza zitshintshiwe. /api/ inesiqendu esipheleleyo.
Iimodeli ezimkelweyo zigcina umbhalo wakho kwi-GPU yethu kwaye ziyisuse emva kokuba unxulumano lubuyiselwe. I-Premium idlula nge-DPA. Asiqeqeshi kwingeniso lakho.
I-Sub-100ms yombhalo omfutshane kwi-self-hosted, 100-500ms kwi-premium. I-batch calls scale roughly linearly — 1,000 chunks complete in 2-10 seconds.
Ewe - Free.ai inika ukusetyenziswa korhwebo kweendawo ezifakwe ngaphakathi. Yenza ukukhangela kophuhliso, i RAG iinqanawa, inkqubo yocebiso ngaphandle kwe royalty nganye ye-vector.