B2B AI API — pricing and quick start

One endpoint for Claude, GPT, Gemini, image and video models — about 2× cheaper than going to each vendor directly. No subscription: you top up a wallet and pay per call at the prices below.

Live prices

ModelTypePrice
claude-opus-5Claude Opustext$2.20 / 1M input tokens
claude-sonnet-4.6Claude Sonnettext$1.32 / 1M input tokens
claude-haiku-4.5Claude Haikutext$0.440 / 1M input tokens
gpt-5.4GPTtext$1.10 / 1M input tokens
gpt-5.4-miniGPT minitext$0.329 / 1M input tokens
nano-banana-2Nano Banana (Gemini)image$0.027 / request
nano-banana-proNano Banana Proimage$0.029 / request
gpt-image-2GPT Imageimage$0.027 / request
kling-v3Klingvideo$0.081 / second of video
seedance-2.0-fastSeedance Fastvideo$0.245 / second of video

Prices track the gateway price list automatically (a fixed markup over the market). Full machine-readable list for every model: GET https://zerocoder.com/api/v1/models

Quick start

  1. Sign in and open the Studio — your API keys live in the account section.
  2. Top up the API wallet (separate from studio credits; minimum first top-up applies).
  3. Call the API — it is drop-in compatible with the Anthropic Messages format:
curl https://zerocoder.com/api/v1/messages \
  -H "x-api-key: YOUR_KEY" \
  -H "content-type: application/json" \
  -d '{"model": "claude-sonnet", "max_tokens": 300,
       "messages": [{"role": "user", "content": "Hello!"}]}'

Model aliases: claude-opus, claude-sonnet, claude-haiku, gpt, gpt-mini, nano-banana, gpt-image and more — see the models endpoint above.

Get an API key

Smart routing — model: "smart"

Don't want to pick a model per request? Pass "smart" (alias "auto") and the router chooses one for each call based on task complexity: simple work goes to fast, inexpensive models, hard work goes to the top tier — so you never overpay for easy calls. Direct model ids keep working exactly as before.

The router is a deterministic zero-latency heuristic — no extra cost, no extra network hop, and the same request always lands in the same tier. It scores each request by: input length, code blocks and programming keywords, reasoning/analysis markers, math, long-form writing, strict output formats (JSON/schema), and dialog length; simple mechanical tasks (translate, fix typos, summarize, classify, extract) lower the score. The total maps to one of three tiers:

TierModelTypical tasks
liteGPT minitranslations, cleanups, extraction, classification, short answers
standardGPTeveryday content: emails, product copy, posts, summaries with nuance
heavyClaude Opuscode, deep analysis, long-form writing, multi-step reasoning
curl https://zerocoder.com/api/v1/messages \
  -H "x-api-key: YOUR_KEY" \
  -H "content-type: application/json" \
  -d '{"model": "smart", "max_tokens": 400,
       "messages": [{"role": "user", "content": "Refactor this function and explain the changes: …"}]}'

Billing is always at the routed model's price (see the table above). The response tells you exactly what ran: the model field carries the actual model id, and smart_routing carries { tier, score, signals } so you can log and audit routing decisions.

FAQ

Why is it cheaper than going direct?

We aggregate volume across one gateway contract and add a fixed markup — that lands about 2× below list prices of the vendors.

Is there a minimum contract?

No. Wallet top-up, pay per call. The wallet is separate from studio credits so your bill reconciles in one currency.

What about rate limits and uptime?

Requests fail over between providers where the model allows it; your key's usage and spend are visible in the account section.

Do you store my prompts?

Requests are proxied for billing and abuse control; we can sign an NDA and discuss data handling for enterprise volumes.

Need volume pricing or an integration done for you?

Book a free consultation