BGE-M3
Free.ai (self-hosted)
·
embeddings
·
~100 tohu i ia call
Ko te BGE-M3 he a tauira whakahuahua i hangaia e te BAAI. He kaha rawa i te Text embeddings, semantic search. I whakanohoia ki ngā GPUs Free.ai - e haere ana i te wātea i runga i tō tātau pūpū tohu i ia rā (100 ngā tohu i ia whakarongo). I tukua i raro i te MIT — i whakaaetia te whakamahinga hokohoko i runga i te Free.ai.
Ka whakamahia mā te API
API REST OpenAI-hōia. Whakana tētahi kī, ā, ka karangatia tēnei tauira i roto i ngā takirua.
curl -X POST https://api.free.ai/v1/image/generate/ \
-H "Authorization: Bearer sk-free-..." \
-H "Content-Type: application/json" \
-d '{"model":"bge-m3","prompt":"your prompt here"}'E pā ana ngā pātai
Ka tahuri te BGE-M3 i te kupu ki tētahi rarangi mārō (he rārangi o ngā pū) e mau ai te tikanga. Ka whakamahia mō te rapu ā-waha, te whakarōpū, te whakapuaki, te whakawhanaketanga whakarei-whakarei (RAG), me ētahi atu mahi e ōrite ana tēnei kupu ki taua kupu" ngā take.
Ko ngā ahu tūturu ko te 384, 768, 1024, 1536 rānei e ai ki te tauira. BGE-M3 e tuku ana i te 1024-dim; E tuku ana a OpenAI Ada i te 1536. Kei roto i te urupare API te ahu kia kōwhiri ai e tōna raupapa DB te taupū tika.
Ko ngā tauira whakahua hōu (tae atu ki te nuinga o ngā kōwhiringa i runga i te Free.ai) kua whakaakona ki ngā reo 100+ me ngā mahi whakaora reo — rapu i te reo Ingarihi, ōrite i ngā tuhinga i te reo Pāniora.
512 ki te 8,192 ngā tohu i runga anō i te tauira. Ka whakaitihia ngā tāuru roa ake — ka whakaitihia ngā tuhinga roa ki ngā wāhanga i mua i te whakatūnga.
E haere ana a BGE-M3 ki a tātau ake GPUs, ā, ko tētahi o ngā utauta iti rawa - tata ki ngā tohu ~100 i ia whakapānga i tangohia mai i tō tātau pūrere wātea i ia rā. $5 = 200K ngā tohu.
He — POST he rārangi aho ki te /v1/embeddings/ ā, ka hoki mai te BGE-M3 ki tētahi rārangi raupapa i te raupapa ōrite. Ko te rahi o te rōpū tae atu ki te 2,048 i ia tono.
L2-whakatika-whakatika e te pūnaha - te ōritetanga o te kōaro = hua pito. Whakapā atu 'whakatika = hē` mēnā e hiahiatia ana e koe ngā rarangi whakakore mō tētahi ine tawhiti rerekē.
He aha — Pinecone, Weaviate, Qdrant, Chroma, pgvector, FAISS, LanceDB. BGE-M3 ka hoki mai i ngā pūrere JSON noa iho; Kāore anō te DB kia kite i te tauira.
He — POST ki /v1/embeddings/ me te tauira "BGE-M3". He āhua urupare OpenAI-hōia, nā reira ko ngā pūranga kaiuru tīariari e mahi ana i te kore huri. He tohutoro katoa te /api/.
Ko ngā tauira ā-whāinga e pupuri ana i tō tātau kupu i runga i a tātau GPUs, ā, ka tangohia i muri i te hokinga o te whakarongo. Ka haere te utu mā te DPA. Kāore e whakaakona e tātau i ōna tāutanga.
Sub-100ms mō te kupu poto i runga i te whakanohotanga, 100–500ms i runga i te utu. Ko ngā whakapāpāho whakatōpū e whakatōpū ana i te paerangi — e 1,000 ngā wāhanga e oti ana i roto i ngā wāhanga 2–10.
He — Free.ai e whakaae ana ki te whakamahi hokohoko o ngā whakatūnga. Hanganga i te rapu whakanao, ngā pūnaha RAG, ngā pūnaha whakawhiwhinga me te kore utu-raupapa.