KeyMart — AI Inference Platform

The Platform

A managed two-sided inference market

We aggregate both official API channels and authorized third-party providers with enterprise agreements. Multi-source averaging delivers below-official pricing while maintaining equivalent SLA through redundancy.

For Developers

All models, one API key, pay-as-you-go

Single OpenAI-compatible endpoint for 25+ models
Top up balance from $5 — pay per token consumed
Smart routing to the best-priced verified provider
Session affinity for KV-cache continuity
Automatic failover in <100ms on throttle or outage
Real-time balance, usage logs, and per-request billing
Works with Cursor, Claude Code, Aider, Windsurf, any OpenAI SDK
99.9% uptime SLA backed by multi-provider redundancy

For Providers

Monetize spare inference capacity

List any model, set your own per-token pricing
Platform routes qualified developer traffic to you
Payouts minus transparent volume-tiered service fee
0% fee for first 7 days — no risk to try
Compliance review + SLA audit before activation
Automated quality monitoring — protect your reputation
Provider Agreement covers data handling & upstream ToS
Self-service withdrawal or scheduled settlement

from openai import OpenAI

client = OpenAI(
    base_url="https://api.keymart.ai/v1",
    api_key="sk-km-your-key-here"
)

# Routed to best available verified provider
response = client.chat.completions.create(
    model="claude-opus-4.7",  # or "gpt-5.5", "deepseek-v4", etc.
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True
)

for chunk in response:
    print(chunk.choices[0].delta.content, end="")

curl https://api.keymart.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-km-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'

# Works with any model listed in the order book
# OpenAI + Anthropic /v1/messages format both supported

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.keymart.ai/v1',
  apiKey: 'sk-km-your-key-here',
});

const stream = await client.chat.completions.create({
  model: 'deepseek-v4',
  messages: [{ role: 'user', content: 'Hello!' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

# Set environment variables
export ANTHROPIC_BASE_URL=https://api.keymart.ai
export ANTHROPIC_API_KEY=sk-km-your-key-here

# Claude Code will use KeyMart automatically
# Native /v1/messages format — no adapter needed

# Or for Cursor / Windsurf / Aider:
export OPENAI_BASE_URL=https://api.keymart.ai/v1
export OPENAI_API_KEY=sk-km-your-key-here

# Any model from the order book is available
# claude-opus-4.7 | gpt-5.5 | gemini-3.1-pro | etc.

Verified providers.One unified API.Below-official pricing.

All models, one API key, pay-as-you-go

Monetize spare inference capacity

Developer Login

Developer Access

Verified providers.
One unified API.
Below-official pricing.