v2.4 — Llama 4 & Qwen3 Now Available

FORGEAPI

ONE KEY. EVERY MODEL. ZERO DOWNTIME.

Route to 20+ AI providers through a single OpenAI-compatible endpoint. Smart failover. Cost optimization. 30% cheaper than direct.

Get API Key — Free View Docs

20+

PROVIDERS

99.99%

UPTIME

~30%

SAVINGS

<50ms

ROUTING

LIVE ROUTER TELEMETRY

HEALTHY

REQUEST

model: claude-4.6-opus

ACCEPTED

ROUTING STRATEGYbalanced

PRIMARY

Anthropic

FAILOVER

Google

INCIDENT SIMULATION

OpenAI latency spike detected. Rerouting to lower-cost healthy path.

RESULTSAVED 31%

200 OK • 842 tokens • 412 ms total

PROVIDER NETWORK

ROUTE ACROSS THE FULL MODEL LAYER

Access frontier models, ultra-fast inference, and specialty infrastructure through one endpoint, one bill, and one API key.

INTEGRATION

THREE STEPS. THAT'S IT.

STEP 01

GET YOUR KEY

forge_sk_live_7f3a...x9k2

STEP 02

DROP IN REPLACE

Change one line. Every OpenAI SDK works out of the box.

base_url = "https://api.forgeapi.dev/v1"

STEP 03

SAVE & SCALE

Automatic routing picks the optimal provider. You set the strategy.

cost speed quality

CODE

OPENAI SDKS WORK OUT OF THE BOX

Keep your current tooling. Swap your key, change your base URL, and ForgeAPI handles the provider layer.

curl https://api.forgeapi.dev/v1/chat/completions \
  -H "Authorization: Bearer forge_sk_..." \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-5.2", "messages": [{"role": "user", "content": "Hello"}]}'

from openai import OpenAI

client = OpenAI(
    base_url="https://api.forgeapi.dev/v1",
    api_key="forge_sk_..."
)

response = client.chat.completions.create(
    model="claude-4.6-opus",
    messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)

import OpenAI from 'openai';

const client = new OpenAI({
    baseURL: 'https://api.forgeapi.dev/v1',
    apiKey: 'forge_sk_...',
});

const response = await client.chat.completions.create({
    model: 'gemini-2.5-pro',
    messages: [{ role: 'user', content: 'Hello' }],
});
console.log(response.choices[0].message.content);

CAPABILITIES

PRODUCTION ROUTING INFRASTRUCTURE

PRICING

CHOOSE YOUR CLEARANCE

RECON

/month

1,000 requests/day
3 providers
Community support
Basic analytics

QUESTIONS, ANSWERED

DEPLOY

START BUILDING IN 30 SECONDS

Free tier. No credit card. 1,000 requests/day. Every major AI model at your fingertips.

DOCUMENTATION

GETTING STARTED

ForgeAPI provides a single, OpenAI-compatible endpoint that routes to 20+ AI providers.

1. Get your API key

2. Install the OpenAI SDK

pip install openai

3. Make your first request

from openai import OpenAI
client = OpenAI(base_url="https://api.forgeapi.dev/v1", api_key="forge_sk_...")
response = client.chat.completions.create(model="gpt-5.2", messages=[{"role": "user", "content": "Hello!"}])

AUTHENTICATION

All requests require a valid API key in the Authorization header.

Authorization: Bearer forge_sk_your_key_here

MODELS

Model	Provider
gpt-5.2	OpenAI
claude-4.6-opus	Anthropic
gemini-2.5-pro	Google
llama-4-180b	Meta
deepseek-v3.2	DeepSeek

RATE LIMITS

Plan	Requests/day	RPM
Recon	1,000	20
Operator	50,000	500
Command	Unlimited	5,000

ERROR CODES

401 UnauthorizedMissing or invalid API key.

429 Rate LimitedPlan threshold exceeded. Retry with backoff.

503 Provider UnavailableAuto-failover engaged. Retry.