Потоковая передача

if payload == "[DONE]":
    break

chunk = json.loads(payload)
if "error" in chunk:
    raise RuntimeError(chunk["error"]["message"])

curl -N -s https://speshu.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $SPESHU_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "'"$TEXT_MODEL"'",
    "stream": true,
    "messages": [{"role": "user", "content": "Напиши три абзаца про океан."}]
  }'

import json
import os

import requests

response = requests.post(
    "https://speshu.ai/api/v1/chat/completions",
    headers={
        "Authorization": f"Bearer {os.environ['SPESHU_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": os.environ["TEXT_MODEL"],
        "stream": True,
        "messages": [{"role": "user", "content": "Напиши три абзаца про океан."}],
    },
    stream=True,
    timeout=(30, 900),  # таймаут чтения потока — с запасом сверх 15 минут
)
response.raise_for_status()

usage = None
for line in response.iter_lines():
    if not line or not line.startswith(b"data: "):
        continue

    payload = line[len(b"data: "):]
    if payload == b"[DONE]":
        break

    chunk = json.loads(payload)

    if "error" in chunk:
        raise RuntimeError(chunk["error"]["message"])

    if "usage" in chunk:
        usage = chunk["usage"]

    for choice in chunk.get("choices", []):
        print(choice.get("delta", {}).get("content", ""), end="", flush=True)

print()
print("списано:", usage["cost"]["total_cost"], "₽")

import json
import os

import httpx


async def stream() -> str:
    text: list[str] = []
    cost = None

    async with httpx.AsyncClient(timeout=httpx.Timeout(900.0)) as client:
        async with client.stream(
            "POST",
            "https://speshu.ai/api/v1/chat/completions",
            headers={
                "Authorization": f"Bearer {os.environ['SPESHU_API_KEY']}",
                "Content-Type": "application/json",
            },
            json={
                "model": os.environ["TEXT_MODEL"],
                "stream": True,
                "messages": [{"role": "user", "content": "Напиши три абзаца про океан."}],
            },
        ) as response:
            response.raise_for_status()

            async for line in response.aiter_lines():
                if not line.startswith("data: "):
                    continue

                payload = line[len("data: "):]
                if payload == "[DONE]":
                    break

                chunk = json.loads(payload)
                if "error" in chunk:
                    raise RuntimeError(chunk["error"]["message"])

                if "usage" in chunk:
                    cost = chunk["usage"]["cost"]["total_cost"]

                for choice in chunk.get("choices", []):
                    piece = choice.get("delta", {}).get("content", "")
                    if piece:
                        text.append(piece)
                        print(piece, end="", flush=True)

    print(f"\nсписано: {cost} ₽")
    return "".join(text)


import asyncio

asyncio.run(stream())

const response = await fetch("https://speshu.ai/api/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.SPESHU_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: process.env.TEXT_MODEL,
    stream: true,
    messages: [{ role: "user", content: "Напиши три абзаца про океан." }],
  }),
});

if (!response.body) throw new Error("поток пуст");

const reader = response.body.getReader();
const decoder = new TextDecoder();
let buffer = "";
let cost: number | null = null;

while (true) {
  const { done, value } = await reader.read();
  if (done) break;
  buffer += decoder.decode(value, { stream: true });

  let newline: number;
  while ((newline = buffer.indexOf("\n")) >= 0) {
    const line = buffer.slice(0, newline).trim();
    buffer = buffer.slice(newline + 1);

    if (!line.startsWith("data: ")) continue;
    const payload = line.slice(6);
    if (payload === "[DONE]") break;

    const chunk = JSON.parse(payload);
    if (chunk.error) throw new Error(chunk.error.message);
    if (chunk.usage) cost = chunk.usage.cost.total_cost;

    for (const choice of chunk.choices ?? []) {
      process.stdout.write(choice.delta?.content ?? "");
    }
  }
}

console.log(`\nсписано: ${cost} ₽`);