0xinf.
live→24,118 req/s
▸0xinf/edge/v0.3.1stable

From 0x,to ∞.Built for builders.

A drop-in OpenAI-compatible gateway. Route across GPT-4o, Claude 3.5, Gemini, Llama and 100+ more — one key, zero 429s, sub-150ms p50. From hello world to a billion tokens.

Get an API key
no card · $5 free credits·11 regions·soc 2 type ii·47d since incident
console.0xinf.com
Requests / 24h
1.4M+12.4%
Uptime
99.99%30d

Active routes

region: iad1
OP
gpt-4o
OpenAI
142ms99.99%
AN
claude-3.5-sonnet
Anthropic
168ms99.97%
GO
gemini-1.5-pro
Google
121ms99.99%
Spend / weekpay-as-you-go
$42.18this week
MTWTFSS
Live metrics
latency
0ms

p50 edge latency

uptime
0%

12-month uptime

coverage
0+

models supported

drop-in
0line

code change

Why 0xinf

The last AI gateway you'll ever need.

One line of code. Every model. Built for teams who ship fast.

Drop-in replacement

Swap one base_url. Your existing OpenAI SDK code works instantly with 100+ models.

base_url = "api.0xinf.com/v1"

Auto-failover

Intelligent routing detects provider outages in milliseconds. Your users never notice.

12ms failover latency

Unified routing

OpenAI, Anthropic, Google, Meta, Mistral — one endpoint, one API key, one invoice.

5 providers, 1 API

Transparent billing

Per-token pricing with millisecond-level logs. No hidden fees, no surprises.

0% gateway fee under $50
Edge-optimizedGlobal PoPs for lowest latency
SOC 2 Type IIEnterprise-grade security
Developer experience

One line.
Infinite models.

Keep your existing OpenAI SDK. Change one URL. Instantly access Claude, Gemini, Llama, and 100+ more models with automatic failover.

OpenAIAnthropicGoogleMetaMistral+95 more
OpenAI compatible
1"text-muted-foreground italic"># pip install openai
2from openai import OpenAI
3
4client = OpenAI(
5    api_key="0xinf_sk_live_***",
6    base_url="https:">//api.0xinf.com/v1",  # ← only change
7)
8
9resp = client.chat.completions.create(
10    model="claude-3-5-sonnet",  "text-muted-foreground italic"># any model
11    messages=[{"role": "user", "content": "ship it"}],
12)
13print(resp.choices[0].message.content)
api.0xinf.com/v1
42ms p50
Supported models

One API key.
Infinite intelligence.

Access every major frontier and open-source model with automatic failover, streaming, and a unified billing dashboard.

OP
GPT-4o
OpenAIflagship
AN
Claude 3.5 Sonnet
Anthropicreasoning
GO
Gemini 1.5 Pro
Google1M context
ME
Llama 3 70B
Metaopen-source
MI
Mistral Large
Mistralfast
OP
GPT-4o mini
OpenAIlow-cost
AN
Claude 3 Haiku
Anthropiclow-latency
GO
Gemini 1.5 Flash
Googlestreaming
CO
Command R+
CohereRAG
DE
DeepSeek V2
DeepSeekMoE
OP
GPT-4o
OpenAIflagship
AN
Claude 3.5 Sonnet
Anthropicreasoning
GO
Gemini 1.5 Pro
Google1M context
ME
Llama 3 70B
Metaopen-source
MI
Mistral Large
Mistralfast
OP
GPT-4o mini
OpenAIlow-cost
AN
Claude 3 Haiku
Anthropiclow-latency
GO
Gemini 1.5 Flash
Googlestreaming
CO
Command R+
CohereRAG
DE
DeepSeek V2
DeepSeekMoE
100+Models
12Providers
WeeklyNew models
Simple pricing

Pay for what you use.
Nothing more.

No hidden fees. No rate limits. Just transparent, per-token billing.

Free Tier

$0

Perfect for trying out 0xinf

  • 100K tokens free monthly
  • All models included
  • Community support
  • Basic dashboard
Start free
Most popular

Developer

$0to start

Pay-as-you-go for growing teams

  • No monthly commitment
  • 0% fee on first $50
  • All 100+ models
  • Real-time dashboard
  • Webhook integrations
  • Priority support
Get started

Enterprise

Custom

For teams with custom needs

  • Volume discounts
  • Dedicated support
  • Custom SLAs
  • SOC 2 Type II
  • SSO & SAML
  • On-prem deployment
Contact sales

All plans include: No rate limits, real-time analytics, and 24/7 monitoring