API Market · buy all compute in one place

Curated premium models and compute, stable and pay-as-you-go

Models & Public Pricing

Model	Input (per 1k tokens)	Output (per 1k tokens)	vs. official
zhimao-gpuzhimao-gpu	0.0105 pts≈ ¥0.0015 / $0.000208	0.042 pts≈ ¥0.006 / $0.000833	↓ save 94%
Zhimao FastUltra-fast, ideal for lightweight tasks	0.015 pts≈ ¥0.002143 / $0.000298	0.06 pts≈ ¥0.008571 / $0.00119	↓ save 92%
zhimao-localzhimao-local	0.005 pts≈ ¥0.000714 / $0.000099	0.02 pts≈ ¥0.002857 / $0.000397	↓ save 97%
Zhimao SmartBalanced performance, great for general use	0.375 pts≈ ¥0.053571 / $0.00744	1.875 pts≈ ¥0.267857 / $0.037202	↓ save 0%
Zhimao VisionVision understanding, multimodal support	0.125 pts≈ ¥0.017857 / $0.00248	0.25 pts≈ ¥0.035714 / $0.00496	↓ save 59%

Official price comparison · supported models

Official retail reference (USD / 1k tokens)

gpt-4o-mini · $0.00015 / $0.0006
claude-sonnet · $0.003 / $0.015

Need a dedicated machine?

Custom environments, fine-tuning, your own images — rent a dedicated GPU by VRAM / region / price in the compute market.

Rent a GPU →

Get your API key

Base URLhttps://zhimao-tokenhub.onrender.com/v1

Set this as base_url in your OpenAI SDK.

Quick-start examples

Replace base_url and api_key with your actual values and start calling immediately.

Python · openai SDK

from openai import OpenAI

client = OpenAI(
    api_key="zm_live_YOUR_KEY",
    base_url="https://zhimao-tokenhub.onrender.com/v1",
)

response = client.chat.completions.create(
    model="zhimao-fast",
    messages=[{"role": "user", "content": "你好"}],
)
print(response.choices[0].message.content)

cURL

curl https://zhimao-tokenhub.onrender.com/v1/chat/completions \
  -H "Authorization: Bearer zm_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zhimao-fast",
    "messages": [{"role":"user","content":"你好"}]
  }'

Learn more · capabilities / 3-step integration / quickstart

An API for any project

From chat to multimodal — one key and one endpoint for every scenario.

Chat / Text API

One endpoint for GPT, Claude, DeepSeek, Qwen and more. OpenAI-compatible, with streaming and function calling.

StreamingFunction callingLong context100+ models

Vision / Multimodal API

Image understanding, visual Q&A, image and video generation — drive multimodal creation from one API.

Image understandingText-to-imageText-to-videoOCR

Agent / tool orchestration

Tool calls, batch jobs and workflows run on our own box fleet — turning model power into deliverable business loops.

Tool callsBatch jobsWorkflowsCompute backing

Integrate in 3 steps

Start in minutes, no credit card required

Sign up & get free credits

Create an account, generate your API key, and claim free trial credits.

Get API Key →

Update Base URL

Point the API endpoint to our address — OpenAI SDK users change just one line.

Start calling models

Access hundreds of text / image / video models through one unified API.

View code samples →

Why Zhimao Compute · FAQ

Why Zhimao Compute

Lower cost

Save 30-70% vs official and mainstream relays. Pay-as-you-go, no monthly or hidden fees.

Multi-model, one API

Unified access to text / image / video models — switch models without changing code. Model catalog growing.

OpenAI compatible

100% drop-in replacement — migrate by changing one base_url line, keep your SDK.

Own compute backing

Not just aggregation — our box GPU fleet provides supply and fulfillment for better margin and stability.

Developer friendly

Built-in code examples (Python / cURL). Onboard in 5 minutes. Real-time usage dashboard to track spend.

Transparent billing

Real-time per-request pricing. Usage dashboard shows today and 30-day spend by model. No hidden fees.

FAQ

What is the AI API Marketplace and what does it offer?

A unified AI platform providing access to leading text / image / video models through a single API, with OpenAI-compatible integration at lower cost than official providers and mainstream relays.

How is pricing calculated?

Everything is metered in Zhimao Points, no monthly fee or minimum. The table shows CNY/USD equivalents for comparison; volume discounts apply automatically.

How do I integrate the API into my app?

Three steps: 1) sign up and create an API key (no credit card); 2) change base_url to our endpoint; 3) start calling in OpenAI-compatible format. OpenAI users change one line.

Why choose you over model providers directly?

One account and one invoice for hundreds of models — no juggling multiple keys. Plus our own box GPU fleet backs supply for better price and stability.

Are my keys secure?

Keys are stored hashed and salted; plaintext is shown only once at creation. You can generate or revoke keys anytime in the console.

Want to supply compute? I have GPUs · Onboarding guide · I have customers · Reselling