{ "aboutHtml": "

BGE Reranker v2 M3 — BGE-M3 asosida qurilgan ko‘p tilli moslik modeli. U so‘rov va nomzod parchani birga qabul qilib, ularning munosabati uchun ball qaytaradi. BAAI bu nashrni, jumladan xitoycha va inglizcha qidiruvda, ilova mavjud nomzodlar ro‘yxatini aniqlashtirishi kerak bo‘lganda unumli ko‘p tilli qayta tartiblovchi sifatida taqdim etadi.

So‘rov–parcha juftligidan moslik

Cross-encoder ikkala matnni birgalikda qayta ishlab, moslik balli yaratilishidan oldin ularning mazmuni o‘zaro ta’sirlashishiga imkon beradi. Bu matnlarni qayta ishlatiladigan vektor taqqoslashi uchun mustaqil kodlaydigan embedding modelidan farq qiladi. Qayta tartiblovchi chiqishi hujjat vektori yoki yozma javob emas, muayyan juftlik uchun skalyar balldir.

Dastlabki qidiruv mos kelishi mumkin bo‘lgan parchalarni tanlagandan keyin modelni ishlating. Savolni har nomzod bilan juftlab, nomzodlarni natija ballari bo‘yicha tartiblang. Bu cross-encoder’dan katta to‘plamdagi har parchani tekshirishni so‘rash o‘rniga, batafsilroq taqqoslashni boshqariladigan qisqa ro‘yxatga qaratadi.

v2 M3 nashri

Ishlab chiqaruvchi BGE-M3’ni asos model deb ko‘rsatadi va ko‘p tilli qo‘llashni asosiy xususiyat sifatida sanaydi. v2 qayta tartiblovchi oilasida M3 unumli ko‘p tilli variant; Gemma asosidagi va qatlam tanlash variantlari boshqa amalga oshirish tanlovlariga ega alohida modellardir. Bu yerda M3 nomi qayta tartiblovchi asl embedding modelining zich, siyrak va ko‘p vektorli embeddinglarini qaytarishini anglatmaydi.

Bu farq ilovaga kerakli natijani tanlashda muhim. Hujjat to‘plamini ifodalash va qidirish uchun embedding modelini tanlang. Nomzod matn allaqachon mavjud bo‘lib, qaysi parcha joriy so‘rovga eng mosligini aniqlash kerak bo‘lsa, shu qayta tartiblovchini tanlang.

Foydali tartiblanadigan nomzodlarni tayyorlash

Haqiqiy savol va uni talqin qilishga kerakli to‘liq nomzod matnni bering. Parcha ma’nosini belgilasa, sarlavhalar yoki yaqin ta’riflarni saqlang. E’lon qilingan Transformers misolida so‘rov va parcha qisqartirish va max_length=512 bilan birga tokenlanadi. Bu butun model limiti emas, misol sozlamasidir. Savol va kerakli dalilni tanlangan juftlik budjetiga sig‘diring va qisqartirish nimani olib tashlashini tekshiring. So‘rovdagi mahsulotni eslatgan natija ham noto‘g‘ri amal yoki shartga tegishli bo‘lsa, aloqasiz bo‘lishi mumkin; qidiruv sifatini baholashda shu holatlarni farqlaydigan misollar tayyorlang.

Hujjat yordamchisida nomzod identifikatorlarini matni bilan saqlang, savol–parcha juftliklarini baholang va tanlangan asl parchalarni javob berish qadamiga yuboring. Shunda javob modeli manbasiz moslik ballini emas, dalillarni oladi. Bu namunaviy qidiruv jarayoni; qayta tartiblovchining o‘zi tashqi to‘plamni qidirmaydi va javob yozmaydi.

Ballarni o‘qish va normallashtirish

Ishlab chiqaruvchi xom moslik ballarini qaytaradi; yuqoriroq qiymat yaqinroq moslikni bildiradi. Hujjatlashtirilgan ixtiyoriy normallashtirish ballni 0 va 1 orasiga joylashtirish uchun sigmoid funksiyasini qo‘llaydi. Normallashtirilgan ball qulay diapazonga ega, ammo bu o‘zgartirishning o‘zi nomzod to‘g‘ri ekanining kalibrlangan ehtimolini belgilamaydi.

Ish yukingizdan olingan tipik so‘rov–parcha juftliklari bilan tartiblash yoki filtrlash chegaralarini tanlang. Umumiy kalit so‘zlari bor, ammo boshqa savolga javob beradigan parchalarni ham, boshqa ifodali tegishli parchalarni ham kiriting. Bu tanlangan chegarani qidirilayotgan to‘plam uchun mazmunli qiladi.

Ailoot API loyihasida yakuniy kontekst nimani o‘z ichiga olishi va qancha nomzodni unumli tekshirish mumkinligini belgilang. Foydali natija — keyingi ilova qadami uchun yaxshiroq tartiblangan dalillar to‘plami. Bu nashrning moslik chiqishini hujjat embeddinglari, yaratilgan nasr va boshqa qayta tartiblovchilar uchun hujjatlashtirilgan qatlam tanlash boshqaruvlaridan alohida tuting.

", "aboutText": "BGE Reranker v2 M3 — BGE-M3 asosida qurilgan ko‘p tilli moslik modeli. U so‘rov va nomzod parchani birga qabul qilib, ularning munosabati uchun ball qaytaradi. BAAI bu nashrni, jumladan xitoycha va inglizcha qidiruvda, ilova mavjud nomzodlar ro‘yxatini aniqlashtirishi kerak bo‘lganda unumli ko‘p tilli qayta tartiblovchi sifatida taqdim etadi.\n\n### So‘rov–parcha juftligidan moslik\n\nCross-encoder ikkala matnni birgalikda qayta ishlab, moslik balli yaratilishidan oldin ularning mazmuni o‘zaro ta’sirlashishiga imkon beradi. Bu matnlarni qayta ishlatiladigan vektor taqqoslashi uchun mustaqil kodlaydigan embedding modelidan farq qiladi. Qayta tartiblovchi chiqishi hujjat vektori yoki yozma javob emas, muayyan juftlik uchun skalyar balldir.\n\nDastlabki qidiruv mos kelishi mumkin bo‘lgan parchalarni tanlagandan keyin modelni ishlating. Savolni har nomzod bilan juftlab, nomzodlarni natija ballari bo‘yicha tartiblang. Bu cross-encoder’dan katta to‘plamdagi har parchani tekshirishni so‘rash o‘rniga, batafsilroq taqqoslashni boshqariladigan qisqa ro‘yxatga qaratadi.\n\n### v2 M3 nashri\n\nIshlab chiqaruvchi BGE-M3’ni asos model deb ko‘rsatadi va ko‘p tilli qo‘llashni asosiy xususiyat sifatida sanaydi. v2 qayta tartiblovchi oilasida M3 unumli ko‘p tilli variant; Gemma asosidagi va qatlam tanlash variantlari boshqa amalga oshirish tanlovlariga ega alohida modellardir. Bu yerda M3 nomi qayta tartiblovchi asl embedding modelining zich, siyrak va ko‘p vektorli embeddinglarini qaytarishini anglatmaydi.\n\nBu farq ilovaga kerakli natijani tanlashda muhim. Hujjat to‘plamini ifodalash va qidirish uchun embedding modelini tanlang. Nomzod matn allaqachon mavjud bo‘lib, qaysi parcha joriy so‘rovga eng mosligini aniqlash kerak bo‘lsa, shu qayta tartiblovchini tanlang.\n\n### Foydali tartiblanadigan nomzodlarni tayyorlash\n\nHaqiqiy savol va uni talqin qilishga kerakli to‘liq nomzod matnni bering. Parcha ma’nosini belgilasa, sarlavhalar yoki yaqin ta’riflarni saqlang. E’lon qilingan Transformers misolida so‘rov va parcha qisqartirish va max_length=512 bilan birga tokenlanadi. Bu butun model limiti emas, misol sozlamasidir. Savol va kerakli dalilni tanlangan juftlik budjetiga sig‘diring va qisqartirish nimani olib tashlashini tekshiring. So‘rovdagi mahsulotni eslatgan natija ham noto‘g‘ri amal yoki shartga tegishli bo‘lsa, aloqasiz bo‘lishi mumkin; qidiruv sifatini baholashda shu holatlarni farqlaydigan misollar tayyorlang.\n\nHujjat yordamchisida nomzod identifikatorlarini matni bilan saqlang, savol–parcha juftliklarini baholang va tanlangan asl parchalarni javob berish qadamiga yuboring. Shunda javob modeli manbasiz moslik ballini emas, dalillarni oladi. Bu namunaviy qidiruv jarayoni; qayta tartiblovchining o‘zi tashqi to‘plamni qidirmaydi va javob yozmaydi.\n\n### Ballarni o‘qish va normallashtirish\n\nIshlab chiqaruvchi xom moslik ballarini qaytaradi; yuqoriroq qiymat yaqinroq moslikni bildiradi. Hujjatlashtirilgan ixtiyoriy normallashtirish ballni 0 va 1 orasiga joylashtirish uchun sigmoid funksiyasini qo‘llaydi. Normallashtirilgan ball qulay diapazonga ega, ammo bu o‘zgartirishning o‘zi nomzod to‘g‘ri ekanining kalibrlangan ehtimolini belgilamaydi.\n\nIsh yukingizdan olingan tipik so‘rov–parcha juftliklari bilan tartiblash yoki filtrlash chegaralarini tanlang. Umumiy kalit so‘zlari bor, ammo boshqa savolga javob beradigan parchalarni ham, boshqa ifodali tegishli parchalarni ham kiriting. Bu tanlangan chegarani qidirilayotgan to‘plam uchun mazmunli qiladi.\n\nAiloot API loyihasida yakuniy kontekst nimani o‘z ichiga olishi va qancha nomzodni unumli tekshirish mumkinligini belgilang. Foydali natija — keyingi ilova qadami uchun yaxshiroq tartiblangan dalillar to‘plami. Bu nashrning moslik chiqishini hujjat embeddinglari, yaratilgan nasr va boshqa qayta tartiblovchilar uchun hujjatlashtirilgan qatlam tanlash boshqaruvlaridan alohida tuting.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "25505ce7abd41b05496bd18e61c422ed18b6ecd85404b1ce57d0b8382c24d41b", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "uz", "methods": [ { "html": "
•

bge-reranker-v2-m3

BAAI /bge-reranker-v2-m3

Boshlash

BAAI’ning nomzod parchalarni so‘rovga nisbatan tartiblash uchun ko‘p tilli cross-encoder modeli. U har bir so‘rov–parcha juftligini birga o‘qib, moslik ballini qaytaradi; javob yaratmasdan yoki hujjat embeddinglarini hosil qilmasdan qidiruvning qisqa ro‘yxatini aniqlashtiradi.

Qabul qilinadigan kirish ma’lumotlari
Matn
Natijalar
Saralangan natijalar
Model haqida

Bu tavsif model imkoniyatlarini yoritadi. Ailoot da tanlangan usul uchun mavjud kirish, chiqish va parametrlar quyida keltirilgan.

BGE Reranker v2 M3 — BGE-M3 asosida qurilgan ko‘p tilli moslik modeli. U so‘rov va nomzod parchani birga qabul qilib, ularning munosabati uchun ball qaytaradi. BAAI bu nashrni, jumladan xitoycha va inglizcha qidiruvda, ilova mavjud nomzodlar ro‘yxatini aniqlashtirishi kerak bo‘lganda unumli ko‘p tilli qayta tartiblovchi sifatida taqdim etadi.

So‘rov–parcha juftligidan moslik

Cross-encoder ikkala matnni birgalikda qayta ishlab, moslik balli yaratilishidan oldin ularning mazmuni o‘zaro ta’sirlashishiga imkon beradi. Bu matnlarni qayta ishlatiladigan vektor taqqoslashi uchun mustaqil kodlaydigan embedding modelidan farq qiladi. Qayta tartiblovchi chiqishi hujjat vektori yoki yozma javob emas, muayyan juftlik uchun skalyar balldir.

Dastlabki qidiruv mos kelishi mumkin bo‘lgan parchalarni tanlagandan keyin modelni ishlating. Savolni har nomzod bilan juftlab, nomzodlarni natija ballari bo‘yicha tartiblang. Bu cross-encoder’dan katta to‘plamdagi har parchani tekshirishni so‘rash o‘rniga, batafsilroq taqqoslashni boshqariladigan qisqa ro‘yxatga qaratadi.

v2 M3 nashri

Ishlab chiqaruvchi BGE-M3’ni asos model deb ko‘rsatadi va ko‘p tilli qo‘llashni asosiy xususiyat sifatida sanaydi. v2 qayta tartiblovchi oilasida M3 unumli ko‘p tilli variant; Gemma asosidagi va qatlam tanlash variantlari boshqa amalga oshirish tanlovlariga ega alohida modellardir. Bu yerda M3 nomi qayta tartiblovchi asl embedding modelining zich, siyrak va ko‘p vektorli embeddinglarini qaytarishini anglatmaydi.

Bu farq ilovaga kerakli natijani tanlashda muhim. Hujjat to‘plamini ifodalash va qidirish uchun embedding modelini tanlang. Nomzod matn allaqachon mavjud bo‘lib, qaysi parcha joriy so‘rovga eng mosligini aniqlash kerak bo‘lsa, shu qayta tartiblovchini tanlang.

Foydali tartiblanadigan nomzodlarni tayyorlash

Haqiqiy savol va uni talqin qilishga kerakli to‘liq nomzod matnni bering. Parcha ma’nosini belgilasa, sarlavhalar yoki yaqin ta’riflarni saqlang. E’lon qilingan Transformers misolida so‘rov va parcha qisqartirish va max_length=512 bilan birga tokenlanadi. Bu butun model limiti emas, misol sozlamasidir. Savol va kerakli dalilni tanlangan juftlik budjetiga sig‘diring va qisqartirish nimani olib tashlashini tekshiring. So‘rovdagi mahsulotni eslatgan natija ham noto‘g‘ri amal yoki shartga tegishli bo‘lsa, aloqasiz bo‘lishi mumkin; qidiruv sifatini baholashda shu holatlarni farqlaydigan misollar tayyorlang.

Hujjat yordamchisida nomzod identifikatorlarini matni bilan saqlang, savol–parcha juftliklarini baholang va tanlangan asl parchalarni javob berish qadamiga yuboring. Shunda javob modeli manbasiz moslik ballini emas, dalillarni oladi. Bu namunaviy qidiruv jarayoni; qayta tartiblovchining o‘zi tashqi to‘plamni qidirmaydi va javob yozmaydi.

Ballarni o‘qish va normallashtirish

Ishlab chiqaruvchi xom moslik ballarini qaytaradi; yuqoriroq qiymat yaqinroq moslikni bildiradi. Hujjatlashtirilgan ixtiyoriy normallashtirish ballni 0 va 1 orasiga joylashtirish uchun sigmoid funksiyasini qo‘llaydi. Normallashtirilgan ball qulay diapazonga ega, ammo bu o‘zgartirishning o‘zi nomzod to‘g‘ri ekanining kalibrlangan ehtimolini belgilamaydi.

Ish yukingizdan olingan tipik so‘rov–parcha juftliklari bilan tartiblash yoki filtrlash chegaralarini tanlang. Umumiy kalit so‘zlari bor, ammo boshqa savolga javob beradigan parchalarni ham, boshqa ifodali tegishli parchalarni ham kiriting. Bu tanlangan chegarani qidirilayotgan to‘plam uchun mazmunli qiladi.

Ailoot API loyihasida yakuniy kontekst nimani o‘z ichiga olishi va qancha nomzodni unumli tekshirish mumkinligini belgilang. Foydali natija — keyingi ilova qadami uchun yaxshiroq tartiblangan dalillar to‘plami. Bu nashrning moslik chiqishini hujjat embeddinglari, yaratilgan nasr va boshqa qayta tartiblovchilar uchun hujjatlashtirilgan qatlam tanlash boshqaruvlaridan alohida tuting.

Ailoot da mavjud · kirish ma’lumotlari va usullar

Ailoot Ultra · Qayta saralash POST /v1/rerank
Kirish
text
Chiqish
rerank
MaydonTurTalabQiymatlar va cheklovlarStandart
MaydonmodelTurstringTalabMajburiyQiymatlar va cheklovlar
bge-reranker-v2-m3
Standart—
MaydonqueryTurstringTalabIxtiyoriyQiymatlar va cheklovlarStandart—
MaydondocumentsTurarrayTalabMajburiyQiymatlar va cheklovlarStandart—
Usul parametrlari va hayot sikli
ParametrParametr talablari
Parametrcontent_typeParametr talablariapplication/json
ParametrrequiredParametr talablarimodel, documents
Parametrmaterial_typesParametr talablaritext
Parametrmaterial_conditionsParametr talablariHaqiqiy qidiruv so‘rovingizni va dastlabki matn hujjatlarining bo‘sh bo‘lmagan massivini belgilang. Natija indekslari shu dastlabki massivga ishora qiladi.

Hayot sikli: synchronous_result_then_owned_library

GET /v1/media/assets/{asset_id}/content

GET /v1/media/assets

GET /v1/media/assets/{asset_id}

Tariflar

Tarif ma’lumotlari so‘rov imkoniyatlarini qo‘shmaydi. Qo‘llab-quvvatlanadigan parametrlar tanlangan usulda ko‘rsatilgan.

Tezkor rejim
Ko‘rsatilmagan Tezkor rejim tanlanganda
Flex rejimi
Ko‘rsatilmagan flex rejimi tanlanganda
Maksimal fikrlash darajasi
Ko‘rsatilmagan Maksimal fikrlash darajasi uchun standart (default) tuzatish qo‘llanganda

Ailoot tariflari

Sozlamalar bo‘yicha narxlar

Har bir to‘lovni ko‘rsatilgan shartlarga moslang. Har bir muqobil sozlamalar to‘plami uchun bitta tarifdan foydalaning.

SozlamaTariflar jadvaliHisob birligiTo‘lov
SozlamaTokenlar uchun bazaviy tariflarTariflar jadvaliAsosiy tarifHisob birligi—To‘lov
Kirish · Kirish tokenlari — / 1K token
Chiqish · Chiqish tokenlari — / 1K token
Kirish · Keshdan o‘qish — / 1K token
Kirish · Keshga yozish — / 1K token
Kirish · Keshga yozish, 1 soat — / 1K token
SozlamaJami tokenTariflar jadvali—Hisob birligi/ 1K tokenTo‘lov
Kirish · Jami token $0.00004 / 1K token

Ailoot so‘rovdan oldin taxminiy summani band qiladi. Natijadan keyin o‘lchangan sarf asosidagi yakuniy to‘lovni yechadi. Oqimli javob bo‘yicha yakuniy hisob-kitob yakuniy sarf yozuviga asoslanadi. Summani band qilish va to‘lovni hisoblash tartibi →

So‘rov namunalari

Ailoot Ultra · Qayta saralash

set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_QUERY:?Set the actual documented input value}"\nexport AILOOT_QUERY\n: "${AILOOT_DOCUMENTS_JSON:?Set the actual documented input value}"\nexport AILOOT_DOCUMENTS_JSON\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\nrequest={'model': 'bge-reranker-v2-m3', 'query': '${ACTUAL_QUERY}', 'documents': '${ACTUAL_DOCUMENTS}'}\nrequest['query']=os.environ['AILOOT_QUERY']\nrequest['documents']=json.loads(os.environ['AILOOT_DOCUMENTS_JSON'])\nPath('request.json').write_text(json.dumps(request,ensure_ascii=False),encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/rerank \\\n  -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n  --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.\n

Natijalarni olish va saqlash

Ailoot Ultra · Qayta saralash

Sozlangan javob modeli, dastlabki hujjat indekslari, moslik baholari, hujjat satrlari va asl token sarfi mavjud to‘liq dastlabki asl JSON natijasi. Har bir haqiqiy natija va foydalanish maydonini saqlang.

Barcha tayyor natijalarni saqlash

Qaytarilgan har bir to‘liq javob, fayl va kontent qismini saqlang.

To‘liq javob va sarlavhalarni, amalda qaytarilgan barcha natijalarni, dastlab qaytarilgan identifikatorlarni hamda barcha usage maydonlarini saqlang. Yaratish so‘rovini qayta yubormasdan dastlabki operatsiyani tiklang.

Asl javobni tekshirish kerak bo‘lsa

Haqiqiy X-Client-Request-ID va to‘liq javobni saqlang. Egalik qilinadigan dastlabki aktivlar va ularning aniq imzolangan mazmun URL larini o‘qing; natijasi noaniq yaratishni hech qachon qayta yubormang.

Javob maydonlari
MaydonTurMa’nosi
MaydonmodelTurstringMa’nosiManbada sozlangan javobni qayta ishlovchi talab qiladigan aniq javob modeli.
Maydonresults[].indexTurintegerMa’nosiDastlabki kirish hujjatining noyob indeksi.
Maydonresults[].relevance_scoreTurnumberMa’nosiChekli moslik bahosi.
Maydonresults[].documentTurstringMa’nosiUshbu sozlangan javob iste’molchisi talab qiladigan dastlabki matn hujjati.
Maydonusage.total_tokensTurintegerMa’nosiHozirgi qabul qilingan tarif talab qiladigan chekli, manfiy bo‘lmagan asl tokenlar soni.
Xatolar va keyingi qadamlar
JavobShartKeyingi qadam
Javob400 · invalid_request_errorShartJSON tanasi yoki ochiq ko‘rsatilgan kanonik model ID si yaroqsiz.Keyingi qadamJavobni saqlang va ko‘rsatilgan so‘rov maydonini tuzating.
Javob401 · authentication_errorShartAiloot API kaliti yaroqsiz.Keyingi qadamRuxsat berilgan Ailoot API kalitingizdan foydalaning.
Javob403 · permission_errorShartKalit guruhi bu usulga ruxsat bermaydi.Keyingi qadamQaytarilgan xato tafsilotlarini saqlang.
Javob404 · not_found_errorShartTanlangan usul uchun ruxsat etilgan yo‘nalish yo‘q.Keyingi qadamQaytarilgan xato tafsilotlarini saqlang.
Javob422 · price_unavailableShartQabul qilingan tarif zarur kirish miqdorini yoki chegarasini aniqlay olmaydi.Keyingi qadamJavobni saqlang; qabul qilingan tarif talab qiladigan kirish ko‘rinishidan foydalaning.
Javob503 · ShartZarur saqlov, yo‘nalish, usul narxi yoki mablag‘ni band qilish imkoniyati mavjud emas.Keyingi qadamTo‘liq dastlabki javobni saqlang. Noaniq mablag‘ band qilish qayta yuborishga ruxsat bermaydi.
Javob502 · billing_requires_checkShartDastlabki natija, foydalanish ma’lumoti yoki hisob-kitobni tekshirish talab qilinadi.Keyingi qadamHaqiqiy so‘rov ID si va dastlabki javobni saqlang. Mavjud, egalik qilinadigan yozuvni o‘qing; yaratishni qayta yubormang.

Taxalluslar

Model taxalluslari

Ailoot Ultra bge-reranker-v2-m3

Usul yo‘llari taxalluslari

Yuqorida ko‘rsatilgan usul yo‘llaridan foydalaning.

Chaqirish uchun tegishli ID lar

bge-reranker-v2-m3

", "method_identity": { "effective_request_model_id": "bge-reranker-v2-m3", "fmt": "rerank", "http": "POST", "path": "/v1/rerank", "request_model_id": "bge-reranker-v2-m3", "route_label": "Ultra" }, "method_key": "Ultra|rerank|/v1/rerank|bge-reranker-v2-m3", "ordinal": 0, "rawMethod": { "available": null, "cache_read": null, "cache_write": null, "cache_write_1h": null, "contract": { "data": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "errors": [ { "code": null, "condition": "JSON tanasi yoki ochiq ko‘rsatilgan kanonik model ID si yaroqsiz.", "http_status": 400, "recovery": "Javobni saqlang va ko‘rsatilgan so‘rov maydonini tuzating.", "type": "invalid_request_error" }, { "code": null, "condition": "Ailoot API kaliti yaroqsiz.", "http_status": 401, "recovery": "Ruxsat berilgan Ailoot API kalitingizdan foydalaning.", "type": "authentication_error" }, { "code": null, "condition": "Kalit guruhi bu usulga ruxsat bermaydi.", "http_status": 403, "recovery": "Qaytarilgan xato tafsilotlarini saqlang.", "type": "permission_error" }, { "code": null, "condition": "Tanlangan usul uchun ruxsat etilgan yo‘nalish yo‘q.", "http_status": 404, "recovery": "Qaytarilgan xato tafsilotlarini saqlang.", "type": "not_found_error" }, { "code": null, "condition": "Qabul qilingan tarif zarur kirish miqdorini yoki chegarasini aniqlay olmaydi.", "http_status": 422, "recovery": "Javobni saqlang; qabul qilingan tarif talab qiladigan kirish ko‘rinishidan foydalaning.", "type": "price_unavailable" }, { "code": null, "condition": "Zarur saqlov, yo‘nalish, usul narxi yoki mablag‘ni band qilish imkoniyati mavjud emas.", "http_status": 503, "recovery": "To‘liq dastlabki javobni saqlang. Noaniq mablag‘ band qilish qayta yuborishga ruxsat bermaydi.", "type": null }, { "code": null, "condition": "Dastlabki natija, foydalanish ma’lumoti yoki hisob-kitobni tekshirish talab qilinadi.", "http_status": 502, "recovery": "Haqiqiy so‘rov ID si va dastlabki javobni saqlang. Mavjud, egalik qilinadigan yozuvni o‘qing; yaratishni qayta yubormang.", "type": "billing_requires_check" } ], "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [], "unknown": "Haqiqiy X-Client-Request-ID va to‘liq javobni saqlang. Egalik qilinadigan dastlabki aktivlar va ularning aniq imzolangan mazmun URL larini o‘qing; natijasi noaniq yaratishni hech qachon qayta yubormang." }, "outputs": { "all_results": "To‘liq javob va sarlavhalarni, amalda qaytarilgan barcha natijalarni, dastlab qaytarilgan identifikatorlarni hamda barcha usage maydonlarini saqlang. Yaratish so‘rovini qayta yubormasdan dastlabki operatsiyani tiklang.", "family": "rerank", "fields": [ { "description": "Manbada sozlangan javobni qayta ishlovchi talab qiladigan aniq javob modeli.", "name": "model", "type": "string" }, { "description": "Dastlabki kirish hujjatining noyob indeksi.", "name": "results[].index", "type": "integer" }, { "description": "Chekli moslik bahosi.", "name": "results[].relevance_score", "type": "number" }, { "description": "Ushbu sozlangan javob iste’molchisi talab qiladigan dastlabki matn hujjati.", "name": "results[].document", "type": "string" }, { "description": "Hozirgi qabul qilingan tarif talab qiladigan chekli, manfiy bo‘lmagan asl tokenlar soni.", "name": "usage.total_tokens", "type": "integer" } ], "result": "Sozlangan javob modeli, dastlabki hujjat indekslari, moslik baholari, hujjat satrlari va asl token sarfi mavjud to‘liq dastlabki asl JSON natijasi. Har bir haqiqiy natija va foydalanish maydonini saqlang." }, "parameters": [ { "default_known": false, "enum": [ "bge-reranker-v2-m3" ], "enum_known": true, "name": "model", "required": true, "type": "string" }, { "default_known": false, "enum_known": false, "name": "query", "required": false, "type": "string" }, { "default_known": false, "enum_known": false, "name": "documents", "required": true, "type": "array" } ], "request_schema": { "properties": { "documents": { "items": { "type": "string" }, "minItems": 1, "type": "array" }, "model": { "enum": [ "bge-reranker-v2-m3" ], "type": "string" }, "query": { "type": "string" } }, "required": [ "model", "documents" ], "type": "object" }, "schema_template": { "actual_file_preparation_required": false, "body": { "documents": "${ACTUAL_DOCUMENTS}", "model": "bge-reranker-v2-m3", "query": "${ACTUAL_QUERY}" }, "content_type": "application/json" } }, "input": { "content_type": "application/json", "material_conditions": [ "Haqiqiy qidiruv so‘rovingizni va dastlabki matn hujjatlarining bo‘sh bo‘lmagan massivini belgilang. Natija indekslari shu dastlabki massivga ishora qiladi." ], "material_types": [ "text" ], "required": [ "model", "documents" ] }, "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "method": { "http": "POST", "path": "/v1/rerank" }, "output": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "rerank", "fields": [ { "description": "Exact response model required by the configured Source consumer.", "name": "model", "type": "string" }, { "description": "Unique original input document index.", "name": "results[].index", "type": "integer" }, { "description": "Finite relevance score.", "name": "results[].relevance_score", "type": "number" }, { "description": "Original text document required by this configured response consumer.", "name": "results[].document", "type": "string" }, { "description": "Finite nonnegative native token count required by the current accepted tariff.", "name": "usage.total_tokens", "type": "integer" } ], "result": "Complete original native JSON result with the configured response model, original document indices, relevance scores, document strings and native token usage. Retain every actual result and usage field." } }, "ctx": null, "declared_pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "/ 1K tokens" }, "label": "total token", "units": [ { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } ], "examples": [ { "code": "set -eu\n: \"${AILOOT_API_KEY:?Set your Ailoot API key}\"\n: \"${AILOOT_QUERY:?Set the actual documented input value}\"\nexport AILOOT_QUERY\n: \"${AILOOT_DOCUMENTS_JSON:?Set the actual documented input value}\"\nexport AILOOT_DOCUMENTS_JSON\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\nrequest={'model': 'bge-reranker-v2-m3', 'query': '${ACTUAL_QUERY}', 'documents': '${ACTUAL_DOCUMENTS}'}\nrequest['query']=os.environ['AILOOT_QUERY']\nrequest['documents']=json.loads(os.environ['AILOOT_DOCUMENTS_JSON'])\nPath('request.json').write_text(json.dumps(request,ensure_ascii=False),encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/rerank \\\n -H \"Authorization: Bearer $AILOOT_API_KEY\" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.\n", "label": "Use actual input and retain the complete result", "language": "bash" } ], "family": "openai", "fast_multiplier": null, "flex_multiplier": null, "fmt": "rerank", "hints": [ "Set your actual query and nonempty array of original text documents. Result indices refer to that original array.", "The minimum example omits top_n, return_documents, instruct and other optional controls.", "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again." ], "http": "POST", "in": null, "input": [ "text" ], "label": "Qayta saralash", "max_out": null, "max_reasoning_effort_multiplier": null, "method_aliases": [], "native_endpoint": "rerank", "out": null, "output": [ "rerank" ], "path": "/v1/rerank", "price_conditions": [ { "label": "Fast mode", "value": "Not specified", "when": "When fast mode is selected" }, { "label": "Flex mode", "value": "Not specified", "when": "When flex mode is selected" }, { "label": "Maximum reasoning effort", "value": "Not specified", "when": "When maximum reasoning effort uses the default adjustment" } ], "pricing": { "billing_mode": "per_request", "image_input_price": null, "image_output_price": null, "per_request_price": null, "unit_prices": { "total_token": 4e-08 } }, "pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "/ 1K tokens" }, "label": "total token", "units": [ { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } ], "request_model_id": "bge-reranker-v2-m3", "route_label": "Ultra", "status_reads": [], "tiers": [], "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" }, { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } } ], "model_id": "bge-reranker-v2-m3", "nativeProvenance": { "input_facts_canonical_sha256": "710327923acd5c3434e867daf9cc7255e0f5d75d87f7cc1b7b6e8a61b2f3e8a8", "input_html_sha256": "24cb7664c06838746bbbda3c67316566180a395c3822b08adcab71b3789d0a2d", "private_audit_sha256": "7406b1fbaaebd6c02810ebdef85f6820330f67a343597e3a6a06e951825aff98", "role": "historical-normalized-input-audit-hashes", "schema": "ailoot-native-model-provenance-v2" }, "page": { "canonical": "https://ailoot.ai/models/uz/bge-reranker-v2-m3/", "description": "BAAI’ning nomzod parchalarni so‘rovga nisbatan tartiblash uchun ko‘p tilli cross-encoder modeli. U har bir so‘rov–parcha juftligini birga o‘qib, moslik ballini qaytaradi; javob yaratmasdan yoki hujjat embeddinglarini hosil qilmasdan qidiruvning qisqa ro‘yxatini aniqlashtiradi.", "locale": "uz", "md_url": "https://ailoot.ai/models/uz/bge-reranker-v2-m3/index.md", "title": "bge-reranker-v2-m3 — Ailoot", "url": "https://ailoot.ai/models/uz/bge-reranker-v2-m3/" }, "rawModel": { "about": "BGE Reranker v2 M3 — BGE-M3 asosida qurilgan ko‘p tilli moslik modeli. U so‘rov va nomzod parchani birga qabul qilib, ularning munosabati uchun ball qaytaradi. BAAI bu nashrni, jumladan xitoycha va inglizcha qidiruvda, ilova mavjud nomzodlar ro‘yxatini aniqlashtirishi kerak bo‘lganda unumli ko‘p tilli qayta tartiblovchi sifatida taqdim etadi.\n\n### So‘rov–parcha juftligidan moslik\n\nCross-encoder ikkala matnni birgalikda qayta ishlab, moslik balli yaratilishidan oldin ularning mazmuni o‘zaro ta’sirlashishiga imkon beradi. Bu matnlarni qayta ishlatiladigan vektor taqqoslashi uchun mustaqil kodlaydigan embedding modelidan farq qiladi. Qayta tartiblovchi chiqishi hujjat vektori yoki yozma javob emas, muayyan juftlik uchun skalyar balldir.\n\nDastlabki qidiruv mos kelishi mumkin bo‘lgan parchalarni tanlagandan keyin modelni ishlating. Savolni har nomzod bilan juftlab, nomzodlarni natija ballari bo‘yicha tartiblang. Bu cross-encoder’dan katta to‘plamdagi har parchani tekshirishni so‘rash o‘rniga, batafsilroq taqqoslashni boshqariladigan qisqa ro‘yxatga qaratadi.\n\n### v2 M3 nashri\n\nIshlab chiqaruvchi BGE-M3’ni asos model deb ko‘rsatadi va ko‘p tilli qo‘llashni asosiy xususiyat sifatida sanaydi. v2 qayta tartiblovchi oilasida M3 unumli ko‘p tilli variant; Gemma asosidagi va qatlam tanlash variantlari boshqa amalga oshirish tanlovlariga ega alohida modellardir. Bu yerda M3 nomi qayta tartiblovchi asl embedding modelining zich, siyrak va ko‘p vektorli embeddinglarini qaytarishini anglatmaydi.\n\nBu farq ilovaga kerakli natijani tanlashda muhim. Hujjat to‘plamini ifodalash va qidirish uchun embedding modelini tanlang. Nomzod matn allaqachon mavjud bo‘lib, qaysi parcha joriy so‘rovga eng mosligini aniqlash kerak bo‘lsa, shu qayta tartiblovchini tanlang.\n\n### Foydali tartiblanadigan nomzodlarni tayyorlash\n\nHaqiqiy savol va uni talqin qilishga kerakli to‘liq nomzod matnni bering. Parcha ma’nosini belgilasa, sarlavhalar yoki yaqin ta’riflarni saqlang. E’lon qilingan Transformers misolida so‘rov va parcha qisqartirish va max_length=512 bilan birga tokenlanadi. Bu butun model limiti emas, misol sozlamasidir. Savol va kerakli dalilni tanlangan juftlik budjetiga sig‘diring va qisqartirish nimani olib tashlashini tekshiring. So‘rovdagi mahsulotni eslatgan natija ham noto‘g‘ri amal yoki shartga tegishli bo‘lsa, aloqasiz bo‘lishi mumkin; qidiruv sifatini baholashda shu holatlarni farqlaydigan misollar tayyorlang.\n\nHujjat yordamchisida nomzod identifikatorlarini matni bilan saqlang, savol–parcha juftliklarini baholang va tanlangan asl parchalarni javob berish qadamiga yuboring. Shunda javob modeli manbasiz moslik ballini emas, dalillarni oladi. Bu namunaviy qidiruv jarayoni; qayta tartiblovchining o‘zi tashqi to‘plamni qidirmaydi va javob yozmaydi.\n\n### Ballarni o‘qish va normallashtirish\n\nIshlab chiqaruvchi xom moslik ballarini qaytaradi; yuqoriroq qiymat yaqinroq moslikni bildiradi. Hujjatlashtirilgan ixtiyoriy normallashtirish ballni 0 va 1 orasiga joylashtirish uchun sigmoid funksiyasini qo‘llaydi. Normallashtirilgan ball qulay diapazonga ega, ammo bu o‘zgartirishning o‘zi nomzod to‘g‘ri ekanining kalibrlangan ehtimolini belgilamaydi.\n\nIsh yukingizdan olingan tipik so‘rov–parcha juftliklari bilan tartiblash yoki filtrlash chegaralarini tanlang. Umumiy kalit so‘zlari bor, ammo boshqa savolga javob beradigan parchalarni ham, boshqa ifodali tegishli parchalarni ham kiriting. Bu tanlangan chegarani qidirilayotgan to‘plam uchun mazmunli qiladi.\n\nAiloot API loyihasida yakuniy kontekst nimani o‘z ichiga olishi va qancha nomzodni unumli tekshirish mumkinligini belgilang. Foydali natija — keyingi ilova qadami uchun yaxshiroq tartiblangan dalillar to‘plami. Bu nashrning moslik chiqishini hujjat embeddinglari, yaratilgan nasr va boshqa qayta tartiblovchilar uchun hujjatlashtirilgan qatlam tanlash boshqaruvlaridan alohida tuting.", "alias_of": null, "aliases": [], "author": "BAAI", "capabilities": { "context_length": 0, "manufacturer_input": [], "manufacturer_output": [] }, "card_routes": [ { "label": "Ultra", "rates": [ { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ], "summary": "Konfiguratsiya narxlari tafsilotlarda to‘liq ko‘rsatilgan." } ], "ctx": null, "formats": [ "rerank" ], "id": "bge-reranker-v2-m3", "input": [ "text" ], "kind": "text", "max_out": null, "methods": [ { "available": null, "cache_read": null, "cache_write": null, "cache_write_1h": null, "contract": { "data": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "errors": [ { "code": null, "condition": "JSON tanasi yoki ochiq ko‘rsatilgan kanonik model ID si yaroqsiz.", "http_status": 400, "recovery": "Javobni saqlang va ko‘rsatilgan so‘rov maydonini tuzating.", "type": "invalid_request_error" }, { "code": null, "condition": "Ailoot API kaliti yaroqsiz.", "http_status": 401, "recovery": "Ruxsat berilgan Ailoot API kalitingizdan foydalaning.", "type": "authentication_error" }, { "code": null, "condition": "Kalit guruhi bu usulga ruxsat bermaydi.", "http_status": 403, "recovery": "Qaytarilgan xato tafsilotlarini saqlang.", "type": "permission_error" }, { "code": null, "condition": "Tanlangan usul uchun ruxsat etilgan yo‘nalish yo‘q.", "http_status": 404, "recovery": "Qaytarilgan xato tafsilotlarini saqlang.", "type": "not_found_error" }, { "code": null, "condition": "Qabul qilingan tarif zarur kirish miqdorini yoki chegarasini aniqlay olmaydi.", "http_status": 422, "recovery": "Javobni saqlang; qabul qilingan tarif talab qiladigan kirish ko‘rinishidan foydalaning.", "type": "price_unavailable" }, { "code": null, "condition": "Zarur saqlov, yo‘nalish, usul narxi yoki mablag‘ni band qilish imkoniyati mavjud emas.", "http_status": 503, "recovery": "To‘liq dastlabki javobni saqlang. Noaniq mablag‘ band qilish qayta yuborishga ruxsat bermaydi.", "type": null }, { "code": null, "condition": "Dastlabki natija, foydalanish ma’lumoti yoki hisob-kitobni tekshirish talab qilinadi.", "http_status": 502, "recovery": "Haqiqiy so‘rov ID si va dastlabki javobni saqlang. Mavjud, egalik qilinadigan yozuvni o‘qing; yaratishni qayta yubormang.", "type": "billing_requires_check" } ], "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [], "unknown": "Haqiqiy X-Client-Request-ID va to‘liq javobni saqlang. Egalik qilinadigan dastlabki aktivlar va ularning aniq imzolangan mazmun URL larini o‘qing; natijasi noaniq yaratishni hech qachon qayta yubormang." }, "outputs": { "all_results": "To‘liq javob va sarlavhalarni, amalda qaytarilgan barcha natijalarni, dastlab qaytarilgan identifikatorlarni hamda barcha usage maydonlarini saqlang. Yaratish so‘rovini qayta yubormasdan dastlabki operatsiyani tiklang.", "family": "rerank", "fields": [ { "description": "Manbada sozlangan javobni qayta ishlovchi talab qiladigan aniq javob modeli.", "name": "model", "type": "string" }, { "description": "Dastlabki kirish hujjatining noyob indeksi.", "name": "results[].index", "type": "integer" }, { "description": "Chekli moslik bahosi.", "name": "results[].relevance_score", "type": "number" }, { "description": "Ushbu sozlangan javob iste’molchisi talab qiladigan dastlabki matn hujjati.", "name": "results[].document", "type": "string" }, { "description": "Hozirgi qabul qilingan tarif talab qiladigan chekli, manfiy bo‘lmagan asl tokenlar soni.", "name": "usage.total_tokens", "type": "integer" } ], "result": "Sozlangan javob modeli, dastlabki hujjat indekslari, moslik baholari, hujjat satrlari va asl token sarfi mavjud to‘liq dastlabki asl JSON natijasi. Har bir haqiqiy natija va foydalanish maydonini saqlang." }, "parameters": [ { "default_known": false, "enum": [ "bge-reranker-v2-m3" ], "enum_known": true, "name": "model", "required": true, "type": "string" }, { "default_known": false, "enum_known": false, "name": "query", "required": false, "type": "string" }, { "default_known": false, "enum_known": false, "name": "documents", "required": true, "type": "array" } ], "request_schema": { "properties": { "documents": { "items": { "type": "string" }, "minItems": 1, "type": "array" }, "model": { "enum": [ "bge-reranker-v2-m3" ], "type": "string" }, "query": { "type": "string" } }, "required": [ "model", "documents" ], "type": "object" }, "schema_template": { "actual_file_preparation_required": false, "body": { "documents": "${ACTUAL_DOCUMENTS}", "model": "bge-reranker-v2-m3", "query": "${ACTUAL_QUERY}" }, "content_type": "application/json" } }, "input": { "content_type": "application/json", "material_conditions": [ "Haqiqiy qidiruv so‘rovingizni va dastlabki matn hujjatlarining bo‘sh bo‘lmagan massivini belgilang. Natija indekslari shu dastlabki massivga ishora qiladi." ], "material_types": [ "text" ], "required": [ "model", "documents" ] }, "lifecycle": { "assets": [ { "http": "GET", "path": "/v1/media/assets" }, { "http": "GET", "path": "/v1/media/assets/{asset_id}" } ], "content": [ { "http": "GET", "path": "/v1/media/assets/{asset_id}/content", "url_source": "exact returned signed content URL" } ], "dedicated_method_polling_or_cancel": false, "mode": "synchronous_result_then_owned_library", "original_POST_adds_task_envelope": false, "status": [], "unknown": "Retain the actual X-Client-Request-ID and complete response. Read the original owned assets and their exact signed content URLs; never resubmit an uncertain creation." }, "method": { "http": "POST", "path": "/v1/rerank" }, "output": { "all_results": "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.", "family": "rerank", "fields": [ { "description": "Exact response model required by the configured Source consumer.", "name": "model", "type": "string" }, { "description": "Unique original input document index.", "name": "results[].index", "type": "integer" }, { "description": "Finite relevance score.", "name": "results[].relevance_score", "type": "number" }, { "description": "Original text document required by this configured response consumer.", "name": "results[].document", "type": "string" }, { "description": "Finite nonnegative native token count required by the current accepted tariff.", "name": "usage.total_tokens", "type": "integer" } ], "result": "Complete original native JSON result with the configured response model, original document indices, relevance scores, document strings and native token usage. Retain every actual result and usage field." } }, "ctx": null, "declared_pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "/ 1K tokens" }, "label": "total token", "units": [ { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } ], "examples": [ { "code": "set -eu\n: \"${AILOOT_API_KEY:?Set your Ailoot API key}\"\n: \"${AILOOT_QUERY:?Set the actual documented input value}\"\nexport AILOOT_QUERY\n: \"${AILOOT_DOCUMENTS_JSON:?Set the actual documented input value}\"\nexport AILOOT_DOCUMENTS_JSON\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\nrequest={'model': 'bge-reranker-v2-m3', 'query': '${ACTUAL_QUERY}', 'documents': '${ACTUAL_DOCUMENTS}'}\nrequest['query']=os.environ['AILOOT_QUERY']\nrequest['documents']=json.loads(os.environ['AILOOT_DOCUMENTS_JSON'])\nPath('request.json').write_text(json.dumps(request,ensure_ascii=False),encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/rerank \\\n -H \"Authorization: Bearer $AILOOT_API_KEY\" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o response.bin\n# Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again.\n", "label": "Use actual input and retain the complete result", "language": "bash" } ], "family": "openai", "fast_multiplier": null, "flex_multiplier": null, "fmt": "rerank", "hints": [ "Set your actual query and nonempty array of original text documents. Result indices refer to that original array.", "The minimum example omits top_n, return_documents, instruct and other optional controls.", "Retain the complete response and headers, every actual output, original returned identifiers and all usage fields. Recover the original operation without submitting creation again." ], "http": "POST", "in": null, "input": [ "text" ], "label": "Qayta saralash", "max_out": null, "max_reasoning_effort_multiplier": null, "method_aliases": [], "native_endpoint": "rerank", "out": null, "output": [ "rerank" ], "path": "/v1/rerank", "price_conditions": [ { "label": "Fast mode", "value": "Not specified", "when": "When fast mode is selected" }, { "label": "Flex mode", "value": "Not specified", "when": "When flex mode is selected" }, { "label": "Maximum reasoning effort", "value": "Not specified", "when": "When maximum reasoning effort uses the default adjustment" } ], "pricing": { "billing_mode": "per_request", "image_input_price": null, "image_output_price": null, "per_request_price": null, "unit_prices": { "total_token": 4e-08 } }, "pricing_profiles": [ { "dimensions": { "regime": "base" }, "label": "Base token rates", "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" } ] }, { "dimensions": { "quoted_unit": "/ 1K tokens" }, "label": "total token", "units": [ { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } ], "request_model_id": "bge-reranker-v2-m3", "route_label": "Ultra", "status_reads": [], "tiers": [], "units": [ { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "input_price", "label": "Input tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "output_price", "label": "Output tokens", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "output_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "out", "unit": "output_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_read_price", "label": "Cache read", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "cached_input_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "cached_input_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_price", "label": "Cache write", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "generic_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "generic_cache_creation_token" }, { "amount": null, "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "cache_write_1h_price", "label": "Cache write, 1 hour", "native_amount_usd_per_unit": null, "native_quantity": 1, "native_unit": "one_hour_cache_creation_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "one_hour_cache_creation_token" }, { "amount": "0.00004000", "amount_unit": "usd_per_thousand_tokens", "billing_input": true, "estimate_only": false, "key": "total_token", "label": "total token", "native_amount_usd_per_unit": "0.00000004", "native_quantity": 1, "native_unit": "total_token", "public_unit_label": "/ 1K tokens", "quantity": 1000, "quantity_is_estimate": false, "rate_role": "typed_native_retail_rate", "side": "in", "unit": "total_token" } ] } ], "model_aliases": [ { "canonical_model_id": "bge-reranker-v2-m3", "request_model_id": "bge-reranker-v2-m3", "route_label": "Ultra" } ], "name": "bge-reranker-v2-m3", "price": { "in": null, "in_label": null, "out": null, "out_label": null, "sort_in": null, "sort_out": null, "varies": true }, "short": "BAAI’ning nomzod parchalarni so‘rovga nisbatan tartiblash uchun ko‘p tilli cross-encoder modeli. U har bir so‘rov–parcha juftligini birga o‘qib, moslik ballini qaytaradi; javob yaratmasdan yoki hujjat embeddinglarini hosil qilmasdan qidiruvning qisqa ro‘yxatini aniqlashtiradi." }, "renderer_schema": { "method_order": "Original DTO methods order; every exact route/format/path/caller survives.", "prices": "Original public DTO fields and original native display HTML; no flattened or normalized offers." }, "representation": "exact-public-dto-and-native-renderer-html", "representationNotes": { "current_reference": { "about_json_pointer": "/aboutHtml", "methods": [ { "contract_json_pointer": "/methods/0/rawMethod/contract", "html_json_pointer": "/methods/0/html", "method_ordinal": 0 } ], "role": "current-semantic-reference" }, "raw_input": { "examples_role": "historical-input-not-current-ready-examples", "json_pointers": [ "/rawModel", "/methods/0/rawMethod" ], "role": "historical-owner-input" }, "schema": "ailoot-native-model-representations-v1" }, "schema": "ailoot-static-model-facts-v1", "schema_version": 1, "shortText": "BAAI’ning nomzod parchalarni so‘rovga nisbatan tartiblash uchun ko‘p tilli cross-encoder modeli. U har bir so‘rov–parcha juftligini birga o‘qib, moslik ballini qaytaradi; javob yaratmasdan yoki hujjat embeddinglarini hosil qilmasdan qidiruvning qisqa ro‘yxatini aniqlashtiradi.", "source": { "audit_sha256": "2e00086bcbfc28e64d0cebe97aa47c17d8d5618963027f8ffb9dc6123d7d25b1", "bridge_map_sha256": "1125275768b9636fc8461bc7da5061239a81ddfe508e2557843ca42c58ab0036", "raw_model_json_sha256": "bef2eedea78463fd7eb2ba7867f1a167aec5ec45a667c64ab815d8ebadf5f080", "renderer_sha256": "aa8a0fca1bd770c042b64987512b6187fca2067aea3ddc877f68b163d80e7df6" }, "sourceRevision": "ab7e3daffbafa25502e8254de546537645f2eed0ed5e2315173d6adf6e39a351" }