What developers build on a $0-inference API

# Continue · ~/.continue/config.json
{
  "models": [{
    "title": "ZeroLimitAI",
    "provider": "openai",
    "model": "auto",
    "apiBase": "https://www.zerolimitai.com/api/v1",
    "apiKey": "zlai_your_key"
  }]
}

from openai import OpenAI
client = OpenAI(base_url="https://www.zerolimitai.com/api/v1", api_key="zlai_your_key")

def reply(history: list[dict], user_text: str) -> str:
    r = client.chat.completions.create(
        model="auto",
        messages=[{"role": "system", "content": DOCS_SYSTEM_PROMPT}, *history,
                  {"role": "user", "content": user_text}],
    )
    return r.choices[0].message.content

r = client.chat.completions.create(
    model="auto",
    messages=[
        {"role": "system", "content": "Summarise in 5 bullets. Then list open questions."},
        {"role": "user", "content": document_text[:32000]},
    ],
)
summary = r.choices[0].message.content
print(r.model)  # the model that actually answered — log it

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="auto", base_url="https://www.zerolimitai.com/api/v1", api_key="zlai_your_key")
answer = llm.invoke(f"Answer from these passages only:\n{passages}\n\nQ: {question}")

import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://www.zerolimitai.com/api/v1", apiKey: "zlai_your_key" });

const draft = await client.chat.completions.create({
  model: "auto",
  messages: [{ role: "user", content: `Write a 120-word product description for: ${sku}` }],
});

r = client.chat.completions.create(
    model="auto",
    messages=[
        {"role": "system", "content": 'Reply with JSON only: {"category": string, "sentiment": "pos"|"neg"|"neu"}'},
        {"role": "user", "content": review_text},
    ],
)
data = json.loads(r.choices[0].message.content)