Moonshot

Kimi K2

Moonshot's flagship. Million-token context window enables whole-repo and whole-corpus prompting without retrieval. Slower per token, but very few alternatives at this context length.

Long contextReasoning

Try Kimi K2 →See plans

OpenAI-compatible

import OpenAI from "openai";

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

const res = await ai.chat.completions.create({
  model: "kimi-k2",
  messages: [{ role: "user", content: "Hello" }],
});

Anthropic-compatible

from anthropic import Anthropic

ai = Anthropic(
    api_key=os.environ["OMNIZEN_KEY"],
    base_url="https://api.omnizen.ai",
)

res = ai.messages.create(
    model="kimi-k2",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)

Other models you might like

DeepSeek

DeepSeek Flash

Cheapest path to a competent model. Excellent for chat and tools.

DeepSeek

DeepSeek V4 Pro

Long-context reasoning model. Strong on math, coding, and planning.

Qwen

Qwen Flash

Tiny, quick, and remarkably cheap. Great default for high-volume jobs.

Start with $1 in free credit.

No card required. Plug into your editor in under a minute.

Start free →Read the docs