Konesans baz / Devlopè

imaj, odyo ak embeddings

Jeneratè yo soti nan app a, ak blòk bati pou rechèch: foto, videyo, pale, transkripsyon ak embeddings, chak chaje pou chak itilizasyon soti nan balans ou.

Paj sa a se machin-tradui pou konvenyans. Orijinal angle a se vèsyon ki aplike a.

Kreye imaj

Li fè youn a kat foto soti nan yon prompt ak retire yo lè yo te fè. Li se menm seri a nan jeneratè kòm ?image Nan aplikasyon an.

POST https://nymbot.ai/api/v1/images/generations — bezwen yon kle API, oswa Yon Lightning peman sou demann.

IDS vini nan GET /api/v1/models?type=imageYon jeneratè estanda, @cf/black-forest-labs/flux-1-schnell, koute yon flat 5 kredi estanda (50 sats) yon foto. Chak lòt jeneratè pase kredi Pro nan pri pwòp li yo, ak frè a nòmal ak marje, kòm nan App nan. n Pwodwi pou foto n Yon jaden yon jeneratè pa pran se chita anvan refize; jeneratè a estanda sèvi ak prompt sèlman.

resolution se pa itilize: chak jeneratè fè gwosè li yo. Avèk url, chak imaj se uploade nan yon òdinatè piblik Blossom ak ou jwenn lyen li (pa gen kèk jeneratè sèlman retire yon lyen pwòp yo, ki se pase sou). Tout moun ki gen lyen an ka louvri li epi li pa ka retire ankò; wè Ki sa ki nan API ka wèavèk b64_json imaj la reponn nan repons la epi yo pa janm ankouraje. Yon imaj voye kòm yon data: URL nan image_url se tcheke nan menm fason ki jan yon Edit nan ak uploaded nan Blossom an premye pou jeneratè a ka retire li; li se retire ankò si demann an pa t 'fwa.

Si kèk nan n imaj manke, ou jwenn moun ki te travay, yo se peye sèlman pou moun (plis nenpòt ki founisè chaje malgre manke), ak nymbot.failed di ki kantite manke. Si tout manke, demann an retire 502.

Reply

{
  "created": 1790726400,
  "model": "nano-banana",
  "cost": 0.2126,
  "data": [
    {
      "url": "https://blossom.example/5c0e4b7d3a91f2e8...c4.png",
      "content_type": "image/png"
    }
  ],
  "nymbot": { "balance": "pro", "charged_credits": 1.817, "charged_sats": 181.7, "balance_credits": 410.608, "balance_sats": 41060.8 }
}

cURL

curl https://nymbot.ai/api/v1/images/generations \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana",
    "prompt": "A lighthouse at dusk, oil painting",
    "size": "16:9"
  }'

Python

import os
import requests

res = requests.post(
    "https://nymbot.ai/api/v1/images/generations",
    headers={"Authorization": "Bearer " + os.environ["NYMBOT_API_KEY"]},
    json={"model": "nano-banana", "prompt": "A lighthouse at dusk, oil painting", "size": "16:9"},
)
res.raise_for_status()
print(res.json()["data"][0]["url"])

JavaScript

const res = await fetch("https://nymbot.ai/api/v1/images/generations", {
  method: "POST",
  headers: {
    "Authorization": "Bearer " + process.env.NYMBOT_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ model: "nano-banana", prompt: "A lighthouse at dusk, oil painting", size: "16:9" }),
});
const body = await res.json();
console.log(body.data[0].url);

Edite imaj

Chanje yon imaj ou upload, apre yon rekòmandasyon. Sa a se Endpoint edit nan OpenAI, se konsa OpenAI SDKs ' images.edit travay kòm li se. Se sèlman Pro jeneratè ak edit nan lis modèl aksepte li.

POST https://nymbot.ai/api/v1/images/edits — bezwen yon kle API, oswa Yon Lightning peman sou demannCorps la se multipart/form-datajiska 32 MB.

imaj la ou upload se tcheke pou yo dwe yon byen fòme PNG, JPEG, WebP oswa GIF, voye sou kòm kalite reyèl li yo, ak mete sou yon òdinatè Blossom piblik, paske jeneratè yo retire entèlijans yo pa lyen. Trete li kòm piblik. Si editè a échoue ak pa gen anyen se chaje, kopi a òdinatè se retire ankò. Repons la gen fòm la menm jan ak pou jenerasyon. Pri a se pri a jeneratè a, ki pou kèk jeneratè gen ladan yon frè pou imaj la entèlijans.

Reply

{
  "created": 1790726460,
  "model": "nano-banana",
  "cost": 0.2126,
  "data": [{ "b64_json": "iVBORw0KGgoAAAANSUhEUgAA..." }],
  "nymbot": { "balance": "pro", "charged_credits": 1.817, "charged_sats": 181.7, "balance_credits": 408.791, "balance_sats": 40879.1 }
}

cURL

curl https://nymbot.ai/api/v1/images/edits \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -F model=nano-banana \
  -F [email protected] \
  -F prompt="Make it a snowy evening"

Python

import base64, os
from openai import OpenAI

client = OpenAI(base_url="https://nymbot.ai/api/v1", api_key=os.environ["NYMBOT_API_KEY"])

with open("photo.png", "rb") as f:
    result = client.images.edit(model="nano-banana", image=f, prompt="Make it a snowy evening")

with open("edited.png", "wb") as out:
    out.write(base64.b64decode(result.data[0].b64_json))

JavaScript

import fs from "node:fs";
import { writeFile } from "node:fs/promises";
import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://nymbot.ai/api/v1", apiKey: process.env.NYMBOT_API_KEY });

const result = await client.images.edit({
  model: "nano-banana",
  image: fs.createReadStream("photo.png"),
  prompt: "Make it a snowy evening",
});
await writeFile("edited.png", Buffer.from(result.data[0].b64_json, "base64"));

Videyo

Yon videyo pran minit, se konsa li kouri nan background la. Ou soumèt li, jwenn yon id imedyatman, ak mande sou id la jiskaske li se fini. Chak videyo jeneratè pase balans la Pro.

POST https://nymbot.ai/api/v1/videos — bezwen yon kle API, oswa Yon Lightning peman sou demannRetounen 202.

Pri a se fixe lè ou soumèt, soti nan jeneratè a, rezolisyon ak longè, se konsa li chaje soti nan salè Pro ou tanpri ke founisè a aksepte travay la, ak montre kòm cost, estimated_cost epi nymbot. Si founisè a refize travay la, pa gen anyen se chaje. Si render la pita manke, oswa clip la fini pa ka sove ak founisè a pa te chaje pou li, chaj la se refize yon fwa, otomatikman. Yon render ki toujou ap ale apre yon èdtan se kite, ak kenbe chaj li yo, paske founisè a chaje yon render yon fwa li aksepte li. Yon refwadisman retire kredi a men pa ba kantite lajan an nan kle a.

Retounen (202)

{
  "id": "vid_3e9a51c07b2f4d86e0a1c4f2",
  "object": "video",
  "model": "veo",
  "status": "in_progress",
  "created": 1790726400,
  "seconds": 8,
  "resolution": "1080p",
  "estimated_cost": 3.18,
  "expires_at": 1790812800,
  "cost": 3.18,
  "nymbot": { "balance": "pro", "charged_credits": 31.8, "charged_sats": 3180, "balance_credits": 378.808, "balance_sats": 37880.8 }
}

status Kòmanse kòm in_progressOu se completed ak videyo a deja nan data lè jeneratè a retire li imedyatman.

Tcheke sou yon videyo

GET https://nymbot.ai/api/v1/videos/{id} - bezwen yon kle ki apwopriye nan nym la ki te soumèt li. Yon kle ki te rive nan cap li kapab toujou tcheke.

Li reponn 202 pandan ke videyo a se fè ak 200 Yon fwa li se konplete oswa te fè mal. Demann chak 5 a 10 segonn. Chèk chak mande founisè a yon fwa ak reponn nan yon sèl fwa; li pa janm ap tann pou videyo a. Yon travay se kenbe pou 24 èdtan apre li te soumèt. GET /api/v1/videos lis dènye 50 travay la nan kle a kòm {"object": "list", "data": […]}.

Repons nan (200)

{
  "id": "vid_3e9a51c07b2f4d86e0a1c4f2",
  "object": "video",
  "model": "veo",
  "status": "completed",
  "created": 1790726400,
  "seconds": 8,
  "resolution": "1080p",
  "estimated_cost": 3.18,
  "expires_at": 1790812800,
  "completed_at": 1790726590,
  "data": { "url": "https://blossom.example/9b41d0c2e7a3f658...1e.mp4", "content_type": "video/mp4" },
  "cost": 3.18,
  "nymbot": { "balance": "pro", "charged_credits": 31.8, "charged_sats": 3180 }
}

status se in_progress, completed ou failedYon travay ki pa t 'kapab pote error ({"code": "generation_failed", "message": …}); si li te retounen, cost se 0 ak nymbot Ajoute refunded_credits epi refunded_satsVideyo a fini se sou yon òdinatè piblik Blossom, tankou imaj yo te kreye.

cURL

curl https://nymbot.ai/api/v1/videos \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "veo", "prompt": "Waves rolling onto a beach at sunrise", "duration": 8, "aspect_ratio": "16:9"}'

curl https://nymbot.ai/api/v1/videos/vid_3e9a51c07b2f4d86 \
  -H "Authorization: Bearer $NYMBOT_API_KEY"

Python

import os, time
import requests

headers = {"Authorization": "Bearer " + os.environ["NYMBOT_API_KEY"]}

job = requests.post(
    "https://nymbot.ai/api/v1/videos",
    headers=headers,
    json={"model": "veo", "prompt": "Waves rolling onto a beach at sunrise", "duration": 8, "aspect_ratio": "16:9"},
).json()

while True:
    time.sleep(8)
    status = requests.get("https://nymbot.ai/api/v1/videos/" + job["id"], headers=headers).json()
    if status["status"] in ("completed", "failed"):
        break

print(status["data"]["url"] if status["status"] == "completed" else status["error"])

JavaScript

const headers = {
  "Authorization": "Bearer " + process.env.NYMBOT_API_KEY,
  "Content-Type": "application/json",
};

const job = await (await fetch("https://nymbot.ai/api/v1/videos", {
  method: "POST",
  headers,
  body: JSON.stringify({ model: "veo", prompt: "Waves rolling onto a beach at sunrise", duration: 8, aspect_ratio: "16:9" }),
})).json();

let status;
do {
  await new Promise((r) => setTimeout(r, 8000));
  status = await (await fetch("https://nymbot.ai/api/v1/videos/" + job.id, { headers })).json();
} while (status.status !== "completed" && status.status !== "failed");

console.log(status.status === "completed" ? status.data.url : status.error);

Tèks nan pale

Li lèt tèks nan wout la ak retire dosye a odyo tèt li, pa yon lyen. Li se OpenAI nan tèks Endpoint, se konsa audio.speech.create nan OpenAI SDKs travay.

POST https://nymbot.ai/api/v1/audio/speech — bezwen yon kle API, oswa Yon Lightning peman sou demann.

Vwa a estanda koute yon plat 3 kredi estanda (30 sats) pou chak demann, tankou nan app la. Aura 2 se Pro, pri pa karaktè yo ou voye. Kò a reponn se odyo a, ak yon korespondan Content-Type; pri a se nan X-Nymbot-Cost-Sats header. Si modèl la pale defèt, pa gen anyen chaje.

Reply

HTTP/1.1 200 OK
Content-Type: audio/mpeg
X-Nymbot-Cost-Sats: 4.2
X-Nymbot-Balance-Sats: 41055.6

(audio bytes)

cURL

curl https://nymbot.ai/api/v1/audio/speech \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "aura-2", "voice": "luna", "input": "Your invoice has been paid."}' \
  --output speech.mp3

Python

import os
from openai import OpenAI

client = OpenAI(base_url="https://nymbot.ai/api/v1", api_key=os.environ["NYMBOT_API_KEY"])

audio = client.audio.speech.create(
    model="aura-2",
    voice="luna",
    input="Your invoice has been paid.",
)
audio.write_to_file("speech.mp3")

JavaScript

import { writeFile } from "node:fs/promises";
import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://nymbot.ai/api/v1", apiKey: process.env.NYMBOT_API_KEY });

const audio = await client.audio.speech.create({
  model: "aura-2",
  voice: "luna",
  input: "Your invoice has been paid.",
});
await writeFile("speech.mp3", Buffer.from(await audio.arrayBuffer()));

Modèl odyo ak vwa

Dwa lis piblik pou chwazi yon modèl pale ak vwa, oswa yon modèl transkripsyon. Gen 41 Aura 2 moun ki pale angle ak kat lang MeloTTS.

GET https://nymbot.ai/api/v1/audio/models Pa gen okenn kle bezwen.

GET https://nymbot.ai/api/v1/audio/voices?language=en Pa gen okenn kle bezwen.

Reply (vòt yo)

{
  "object": "list",
  "data": [
    { "id": "luna", "name": "Luna", "provider": "deepgram", "model_id": "aura-2", "models": ["aura-2"], "language": "en", "description": "Aura 2 English voice", "gender": "female", "preview_url": null },
    { "id": "en", "name": "MeloTTS English", "provider": "myshell", "model_id": "@cf/myshell-ai/melotts", "models": ["@cf/myshell-ai/melotts", "melotts"], "language": "en", "description": "English (MeloTTS)", "gender": null, "preview_url": null }
  ]
}

Modèl lis repons {"object": "list", "data": {"tts": […], "stt": […]}}Tout modèl pale bay balans li yo, pri, max_input_chars, response_formats, languages epi default_voice; modèl la transkripsyon bay alyans li yo, max_file_bytes, max_seconds, fòma ak pri pou chak minit.

cURL

curl https://nymbot.ai/api/v1/audio/models
curl "https://nymbot.ai/api/v1/audio/voices?language=en"

Python

import requests

print(requests.get("https://nymbot.ai/api/v1/audio/models").json())
for voice in requests.get("https://nymbot.ai/api/v1/audio/voices", params={"language": "en"}).json()["data"]:
    print(voice["id"], voice["model_id"])

JavaScript

console.log(await (await fetch("https://nymbot.ai/api/v1/audio/models")).json());
const voices = await (await fetch("https://nymbot.ai/api/v1/audio/voices?language=en")).json();
for (const v of voices.data) console.log(v.id, v.model_id);

Diskisyon nan tèks

Konvèti yon anrejistreman nan tèks ak Whisper. Sa a se OpenAI a transkripsyon Endpoint, se konsa audio.transcriptions.create travay nan.

POST https://nymbot.ai/api/v1/audio/transcriptions — bezwen yon kle API, oswa Yon Lightning peman sou demannCorps la se multipart/form-data.

Pou dosye WAV, WebM, MP4 ak Ogg, longè a se li soti nan dosye a anvan anyen kouri, se konsa yon anrejistreman plis pase 30 minit refize nan devan (ak 3 sekondè 'grati). Longè a nan yon dosye WAV vini nan vitès echantiyon li yo, chanèl yo ak gwosè echantiyon. Longè a nan lòt fòma, tankou MP3, pa ka li dwe li nan devan, se konsa yo limite nan gwosè: nan pi ba bitrate Nymbot asume, 8 kbps, 30 minit se 1.8 MB, ak yon dosye pi gwo nan fòma sa a refize ak 413 audio_too_long voye enskripsyon pi long tankou WAV, WebM, MP4 oswa Ogg. Si Whisper Lè sa a rapòte plis pase 30 minit, demann an se refize ak 413 audio_too_long, ak depi Whisper te kouri, 30 minit yo chaje.

Ou peye pou longè a nan odyo a, aranjman jiska dezyèm: pi long nan longè a li soti nan dosye a, longè a rapòte Whisper, ak pi piti gwosè a dosye a pèmèt, nan pri a per-minit nan Whisper a ak frè a nòmal ak marje, ak omwen 0.05 kredi yon demann. Li se peye soti nan balans la estanda lè sa a ka kouvri kenbe, ak soti nan Pro anyen; 402 bay sa ki chak balans ta bezwen. Nan aplikasyon an, transkripsyon yon mesaj vwa ka gratis; sou API a li se toujou chaje. Si Whisper échoue, pa gen anyen chaje.

Reply

{
  "text": "Your invoice has been paid.",
  "usage": { "type": "duration", "seconds": 3 },
  "nymbot": { "balance": "standard", "charged_credits": 0.05, "charged_sats": 0.5, "balance_credits": 120.35, "balance_sats": 1203.5 }
}

text Retire tèks klè, srt epi vtt Retour sous tit, ak verbose_json ajoute lang, tanperati ak segman timed. Pou twa ki pa JSON, pri a se nan tèt yo.

cURL

curl https://nymbot.ai/api/v1/audio/transcriptions \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -F [email protected] \
  -F model=whisper \
  -F response_format=json

Python

import os
from openai import OpenAI

client = OpenAI(base_url="https://nymbot.ai/api/v1", api_key=os.environ["NYMBOT_API_KEY"])

with open("memo.m4a", "rb") as f:
    transcript = client.audio.transcriptions.create(model="whisper", file=f)
print(transcript.text)

JavaScript

import fs from "node:fs";
import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://nymbot.ai/api/v1", apiKey: process.env.NYMBOT_API_KEY });

const transcript = await client.audio.transcriptions.create({
  model: "whisper",
  file: fs.createReadStream("memo.m4a"),
});
console.log(transcript.text);

Tradiksyon nan lang angle

POST https://nymbot.ai/api/v1/audio/translations — bezwen yon kle API, oswa Yon Lightning peman sou demannCorps la se multipart/form-data.

menm jan ak transkripsyon, ak menm jaden an nan language, limit ak pri, men tèks la vin tounen nan lang angle nenpòt ki lang te pale. OpenAI SDK yo rele li ak audio.translations.create.

cURL

curl https://nymbot.ai/api/v1/audio/translations \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -F [email protected] \
  -F model=whisper

Python

import os
from openai import OpenAI

client = OpenAI(base_url="https://nymbot.ai/api/v1", api_key=os.environ["NYMBOT_API_KEY"])

with open("nota.m4a", "rb") as f:
    translation = client.audio.translations.create(model="whisper", file=f)
print(translation.text)

JavaScript

import fs from "node:fs";
import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://nymbot.ai/api/v1", apiKey: process.env.NYMBOT_API_KEY });

const translation = await client.audio.translations.create({
  model: "whisper",
  file: fs.createReadStream("nota.m4a"),
});
console.log(translation.text);

entwodiksyon

Konvèti tèks nan vèktè pou rechèch, klasman ak deduplikasyon. Modèl yo kouri sou Cloudflare ak pase balans la estanda.

POST https://nymbot.ai/api/v1/embeddings — bezwen yon kle API, oswa Yon Lightning peman sou demann.

List la modèl se sous a nan verite pou ki yo disponib ak ki sa yo koute; Google a EmbeddingGemma, ak vètikulè pi kout sou demann, ap parèt isit la tou lè kataloz la gen yon pri pou li. Enbalaj yo mete pou chak token envantè, ak minimòm la 0.05 kredi aplike pou chak demann, pa pou chak envantè, se konsa batch anpil tèks nan yon sèl demann se anpil pi bon mache pase voye yo yon sèl nan yon fwa.

Reply

{
  "object": "list",
  "data": [
    { "object": "embedding", "index": 0, "embedding": [0.0213, -0.0487, 0.0112] }
  ],
  "model": "@cf/baai/bge-m3",
  "usage": { "prompt_tokens": 9, "total_tokens": 9 },
  "nymbot": { "balance": "standard", "charged_credits": 0.05, "charged_sats": 0.5, "balance_credits": 1204.35, "balance_sats": 12043.5 }
}

cURL

curl https://nymbot.ai/api/v1/embeddings \
  -H "Authorization: Bearer $NYMBOT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "@cf/baai/bge-m3", "input": ["Lightning invoices expire.", "Invoices on Lightning have a time limit."]}'

Python

import os
from openai import OpenAI

client = OpenAI(base_url="https://nymbot.ai/api/v1", api_key=os.environ["NYMBOT_API_KEY"])

result = client.embeddings.create(
    model="@cf/baai/bge-m3",
    input=["Lightning invoices expire.", "Invoices on Lightning have a time limit."],
)
print(len(result.data), len(result.data[0].embedding))

JavaScript

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://nymbot.ai/api/v1", apiKey: process.env.NYMBOT_API_KEY });

const result = await client.embeddings.create({
  model: "@cf/baai/bge-m3",
  input: ["Lightning invoices expire.", "Invoices on Lightning have a time limit."],
});
console.log(result.data.length, result.data[0].embedding.length);