model

Gemini Embedding 001 Batch

Markdown · JSON

Gemini Embedding 001 Batch

google/gemini-embedding-001:batch
GooglePembenamanAuto · Ultra

Menukar teks kepada vektor melalui tugas kelompok tak segerak.

Ultra BatchPOST /v1/batches
Input
text
Output
embedding
Konteks
Tidak dinyatakan dalam katalog
Respons maksimum
Tidak dinyatakan dalam katalog
Input
USD 0.00015 / 1K
Output
USD 0.00 / 1K
Parameter kaedah dan kitar hayat
ParameterKontrak
ParameterendpointKontrak/v1/embeddings
ParametermodelKontrakgoogle/gemini-embedding-001:batch
ParameterrequestsKontrakcustom_id + body.input

Kitar hayat: asynchronous

GET /v1/batches/{request_id} · response.id

POST /v1/batches

curl --silent --show-error https://api.ailoot.ai/v1/batches \
  -H "Authorization: Bearer $AILOOT_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Ailoot-Routing-Mode: ultra" \
  --data '{"model":"google/gemini-embedding-001:batch","endpoint":"/v1/embeddings","requests":[{"custom_id":"text-1","body":{"input":"First document"}}]}'

GET /v1/batches/{request_id}

curl --silent --show-error "https://api.ailoot.ai/v1/batches/$BATCH_ID" \
  -H "Authorization: Bearer $AILOOT_API_KEY" \
  -H "X-Ailoot-Routing-Mode: ultra"
JSON
{"about":"Menukar teks kepada vektor melalui tugas kelompok tak segerak.","alias_of":null,"aliases":[],"author":"Google","available":true,"canonical_model":"google/gemini-embedding-001:batch","card_routes":[{"label":"Ultra","rates":[{"amount":0.15,"amount_unit":"usd_per_million_tokens","key":"in","label":"Input","side":"in"}]}],"ctx":null,"formats":["batches"],"id":"google/gemini-embedding-001:batch","input":["text"],"kind":"embedding","max_out":null,"methods":[{"contract":{"data":{"example_body":{"endpoint":"/v1/embeddings","model":"google/gemini-embedding-001:batch","requests":[{"body":{"input":"First document"},"custom_id":"text-1"}]},"outputs":{"family":"embedding","fields":[{"name":"id","type":"string"},{"name":"status","type":"string"},{"name":"results[].custom_id","type":"string"},{"name":"results[].response.body.data[].embedding","type":"number[]"},{"name":"results[].error","type":"object|null"}],"format":"JSON","vector_dimensions":3072},"parameters":[{"enum":["google/gemini-embedding-001:batch"],"name":"model","required":true,"type":"string"},{"enum":["/v1/embeddings"],"name":"endpoint","required":true,"type":"string"},{"min_items":1,"name":"requests","required":true,"type":"batch_item[]"},{"name":"requests[].custom_id","required":true,"type":"string","unique_within_batch":true},{"name":"requests[].body.input","required":true,"type":"string"}]},"hints":[],"input":{"endpoint":"/v1/embeddings","model":"google/gemini-embedding-001:batch","requests":"custom_id + body.input"},"lifecycle":{"create":{"accepted_status":202,"http":"POST","path":"/v1/batches"},"mode":"asynchronous","status":{"http":"GET","id_source":"response.id","path":"/v1/batches/{request_id}"}},"method":{"http":"POST","path":"/v1/batches"},"output":{"family":"embedding","fields":[{"name":"id","type":"string"},{"name":"status","type":"string"},{"name":"results[].custom_id","type":"string"},{"name":"results[].response.body.data[].embedding","type":"number[]"},{"name":"results[].error","type":"object|null"}],"format":"JSON","vector_dimensions":3072}},"ctx":null,"examples":[{"code":"curl --silent --show-error https://api.ailoot.ai/v1/batches \\\n  -H \"Authorization: Bearer $AILOOT_API_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -H \"X-Ailoot-Routing-Mode: ultra\" \\\n  --data '{\"model\":\"google/gemini-embedding-001:batch\",\"endpoint\":\"/v1/embeddings\",\"requests\":[{\"custom_id\":\"text-1\",\"body\":{\"input\":\"First document\"}}]}'","label":"curl · POST /v1/batches","language":"bash"},{"code":"curl --silent --show-error \"https://api.ailoot.ai/v1/batches/$BATCH_ID\" \\\n  -H \"Authorization: Bearer $AILOOT_API_KEY\" \\\n  -H \"X-Ailoot-Routing-Mode: ultra\"","label":"curl · GET /v1/batches/{request_id}","language":"bash"}],"family":"batch","fmt":"batches","hints":[],"http":"POST","img_in":null,"img_out":null,"in":0.15,"input":["text"],"label":"Batch","max_out":null,"native_endpoint":"batches","out":0,"output":["embedding"],"path":"/v1/batches","pricing":{"billing_mode":"token","cache_write_1h_usd_per_token":null,"cache_write_5m_usd_per_token":null,"currency":"USD","input_usd_per_token":1.5e-07,"output_usd_per_token":0,"price_policy":"native_accepted_retail_card","provenance":"native_group_model_pricing","status":"configured","tiers":[{"cache_write_1h_usd_per_token":null,"cache_write_5m_usd_per_token":null,"input_usd_per_token":1.5e-07,"interval_boundary":"min_exclusive_max_inclusive","label":"","min_tokens":0,"null_rate_policy":"inherit_base_rate","output_usd_per_token":0}]},"request_model_id":"google/gemini-embedding-001:batch","route_label":"Ultra","route_tier":"ultra","status_reads":["GET /v1/batches/{request_id}"],"tiers":[],"unit":"token","units":[]}],"model_aliases":[],"name":"Gemini Embedding 001 Batch","price":{"in":0.15,"in_label":null,"out":0,"out_label":null,"sort_in":0.15,"sort_out":0,"varies":false},"public_id":"google/gemini-embedding-001:batch","short":"Menukar teks kepada vektor melalui tugas kelompok tak segerak."}

Soalan

Bagaimanakah saya memilih model untuk tugasan saya?

Kenal pasti dahulu input dan output yang diperlukan: teks, imej, audio atau video. Kemudian bandingkan kaedah API yang serasi, ID model yang tepat atau model yang dirujuk oleh alias pada masa ini, harga dan unit pengebilan. GET /v1/models adalah percuma dan menyenaraikan model yang tersedia untuk kunci anda.

Adakah setiap model menyokong semua kaedah API dan jenis media?

Gunakan kaedah API dan jenis input/output yang disenaraikan pada kad model. Sokongan untuk teks, imej, audio atau video bergantung pada model dan kaedah yang dipilih; semak contoh permintaan yang berkenaan sebelum menghantar data anda.

Apakah perbezaan antara ID model dengan alias?

ID model yang tepat mengenal pasti model tertentu. Alias merujuk kepada model dan model sasarannya boleh berubah. Semak model yang dirujuk oleh alias pada masa ini dalam katalog; gunakan ID yang tepat apabila integrasi anda memerlukan model tertentu.

Bagaimanakah saya membandingkan harga sebelum menghantar permintaan?

Semak kad model untuk harga dan unit pengebilan bagi kaedah API yang dipilih. Bandingkan menggunakan unit dan jenis input/output yang sama; sejarah permintaan menunjukkan penggunaan sebenar dan amaun yang dicaj.

Apakah manfaat langganan?

Diskaun pembelian token untuk penggunaan berkala. Pilih pelan berdasarkan jumlah permintaan; terma dan harga langganan dipaparkan semasa pembelian.

Bagaimanakah kod promosi diskaun sehingga 50% digunakan?

Masukkan kod promosi semasa membeli token. Jumlah akhir dikira semula mengikut terma promosi; kadar diskaun kelihatan sebelum pembayaran.