mxbai-embed-large-v1
Free.ai (self-hosted)
·
embeddings
·
~100 tokens per call
mxbai-embed-large-v1 iku an embedding model dibangun déning mixedbread.ai. Strongest at Semantic search, clustering, similarity.. Self-hosted on Free.ai GPUs — runs free against your daily token pool (100 tokens saben telpon). Released under Apache 2.0 — commercial use permitted on Free.ai.
Takon kang asring diajukake
mxbai-embed-large-v1 ngowahi teks dadi vektor kang padat (daftar angka kang ajeg) kang nyekel makna. Gunakaké kanggo pencarian semantik, clustering, rekomendasi, retrieval-augmented generation (RAG), lan tugas apa wae kang "apa teks iki padha karo teks iku" penting.
Dimènsi ingkang umum inggih punika 384, 768, 1024, utawi 1536 gumantung saking modélipun. BGE-M3 ngetokaken 1024-dim; OpenAI Ada ngetokaken 1536.
Model embedding modern (kajaba pilihan paling ing Free.ai) dilatih ing 100+ basa. Cross-basa retrieval karya - goleki ing Inggris, match dokumen ing Spanyol.
512 nganti 8,192 token gumantung karo modél. Input kang luwih dawa bakal dipotong - chunk dokumen dawa dadi paragraf sadurunge ngembang.
mxbai-embed-large-v1 runs on our own GPUs and is among the cheapest tools - about ~100 tokens per call drawn from your daily free pool. $5 = 200K tokens.
Ya - POST dhaptar strings ka /v1/embeddings/ lan mxbai-embed-large-v1 ngembali dhaptar vektor ing urutan kang padha. Ukuran batch nganti 2,048 saben pitakon.
L2-normalisasi déning pigura pigura - cosinus similarity = titik produk. Ngetokaken 'normalize=false' yèn sampeyan pengin vektor raw kanggo jarak metrik sing beda.
Any - Pinecone, Weaviate, Qdrant, Chroma, pgvector, FAISS, LanceDB. mxbai-embed-large-v1 returns plain JSON floats; the DB never sees the model.
Yes - 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.
Model self-hosted nyimpen teks ing GPU kita lan mbusak sawisé panjaluk balik. Premium ngliwati karo DPA. Kita ora latihan ing input sampeyan.
Sub-100ms kanggé teks cekak ing self-hosted, 100-500ms ing premium. Batch panjaluk skala kira-kira linear - 1,000 chunks rampung ing 2-10 detik.
Ya - Free.ai nyedhiyani panggunaan komersial ing ngembang. Ngwangun panjaluk produksi, pipa RAG, sistem rekomendasi tanpa royalti per-vektor.