LLM API Marketplace is now live

Access Every AI Model
Through One API

GPT, Claude, Gemini, Llama. One key compatible with OpenAI/Anthropic, transparent pricing, and a marketplace where providers earn from real usage.

Start for free Browse models

120M+

API Requests / day

150+

Models available

30+

Active providers

<2000ms

Avg latency

Everything you need to use AI at scale

Built for developers who want flexibility, and for providers who want to monetize.

One Key, Every Model

A single OpenAI-compatible key unlocks GPT, Claude, Gemini and every provider-listed model. Zero SDK changes.

Learn more

Verified Providers

Every provider is reviewed before going live. Real-time uptime, latency, and success-rate metrics are public.

Learn more

Pay Per Token

No subscriptions, no seats. Load your wallet and pay only for the tokens you actually consume.

Learn more

Become a Provider

Host your own inference endpoint, list it on the marketplace and earn on every token buyers send through.

Learn more

Usage Analytics

Token counts, cost breakdown, latency histograms, and full request logs visible in the console in real time.

Learn more

Secure by Design

API keys are SHA-256 hashed. Provider upstream keys are AES-256-GCM encrypted. JWT auth with short-lived tokens.

Learn more

Latest models

Competitive pricing from verified providers.

View all

GPT-5.5

OpenAI

Flagship

Input / 1M

$0.25

Output / 1M

$1.50

GPT-5.4

OpenAI

Latest

Input / 1M

$0.125

Output / 1M

$0.75

GPT-5.4 mini

OpenAI

Value

Input / 1M

$0.0375

Output / 1M

$0.225

GPT-5.3 Codex

OpenAI Codex

Coding

Input / 1M

$0.0875

Output / 1M

$0.70

Claude Opus 4.7

Anthropic

Frontier

Input / 1M

$0.25

Output / 1M

$1.25

Claude Sonnet 4.6

Anthropic

Agentic

Input / 1M

$0.15

Output / 1M

$0.75

Claude Haiku 4.5

Anthropic

Fast

Input / 1M

$0.05

Output / 1M

$0.25

Gemini 3.1 Pro Preview

Google

Multimodal

Input / 1M

$0.10

Output / 1M

$0.60

Gemini 3 Flash Preview

Google

Fast

Input / 1M

$0.025

Output / 1M

$0.15

Gemini 3.1 Flash-Lite

Google

Cheap

Input / 1M

$0.0125

Output / 1M

$0.075

DEDeepSeek V4 Flash

DeepSeek

Best value

Input / 1M

$0.007

Output / 1M

$0.014

MIMistral Large 3

Mistral

Open

Input / 1M

$0.025

Output / 1M

$0.075

OpenAI compatible

Compatible with the AI tools you already use

Use Foza models inside popular editors, CLIs, and agent frameworks with only a base URL and API key.

OpenCode

CLI

Configure Foza as an OpenAI-compatible provider for local workflows.

config
# ~/.config/opencode/config.json
"providers": {
  "foza": {
    "base_url": "https://api.foza.ai/v1",
    "api_key": "sk-foza-xxxx"
  }
}

Cursor

AI Editor

Use it in your AI editor with a custom base URL and Foza model aliases.

config
# Cursor Settings → Models
Override AI Base URL:
  https://api.foza.ai/v1
 
API Key: sk-foza-xxxx
Model:   provider/gpt-4o

Claude Code

CLI Agent

Connect Claude-compatible models or route coding tasks through the gateway.

config
$ ANTHROPIC_BASE_URL=\
    https://api.foza.ai/v1 \
  ANTHROPIC_API_KEY=sk-foza-xxxx \
  claude

Codex CLI

Terminal

Run terminal tasks with one API key and a unified model list.

config
$ OPENAI_API_KEY=sk-foza-xxxx \
  OPENAI_BASE_URL=\
    https://api.foza.ai/v1 \
  codex "Fix the bug in main.ts"
OC

OpenClaw

Agent

Attach the gateway to agent runtimes and switch models without logic changes.

config
# openclaw.config.ts
llm: {
  provider: "openai",
  base_url: "https://api.foza.ai/v1",
  api_key: "sk-foza-xxxx",
  model: "provider/claude-sonnet-4.6"
}
HA

Harness Agent

Workflow

Power agent workflows with cost, token, and request logs built in.

config
# harness-agent.yaml
ai_provider:
  type: openai_compatible
  base_url: https://api.foza.ai/v1
  api_key: ${{ secrets.FOZA_KEY }}
  model: provider/gpt-5.4

Shared base URL

https://api.foza.ai/v1 — drop-in replacement for any OpenAI base URL

Set up now

How it works

From zero to first AI response in under 5 minutes.

01

Create an account

Sign up with email or Google in under 30 seconds.

02

Generate an API key

Create a key in the console and subscribe it to any model.

03

Call the gateway

Use any OpenAI-compatible SDK, just change base URL and key.

04

Monitor and scale

Track spend, tokens, and latency. Top up and keep going.

For Providers

Monetize your
API access

Already paying for GPT, Claude, or running your own models? Turn that into income. Register, list your models with custom pricing, and earn on every token buyers send.

  • Set your own per-token pricing
  • Bring your existing API access
  • Earnings credited to wallet per request
  • Usage analytics per model
  • No infrastructure changes needed
  • Admin approval ensures marketplace quality
Become a provider See the marketplace
Provider Console
Your earnings at a glance
Total Earnings
$1,284.50
Active Models
3 models
Requests (30d)
48,291
Avg Latency
142ms
gpt-4o-mini (provider)Active
$0.15 / 1M input · $0.60 / 1M output
+12.4% vs last week

Ready to get started?

Create a free account, generate an API key, and ship your first request in minutes.

Create free account Sign in