One API for many AI models with IDR billing.
Genfity AI Gateway unifies API keys, credits, quotas, and model routing in one stable endpoint. Pay via QRIS or bank transfer—no USD credit card required.
Supported providers
OpenAI
Anthropic
Google Gemini
Meta
Mistral AI
xAI
Perplexity
Hugging Face
CohereOpenAIAnthropicGoogle GeminiMetaMistral AIxAIPerplexityHugging FaceCohereOpenAIAnthropicGoogle GeminiMetaMistral AIxAIPerplexityHugging FaceCohereWhy Genfity
More than just an AI proxy.
Infrastructure built for Indonesian developers: pay in rupiah, local docs, WIB support team.
One API, all models
Switch between models by changing the model name string. No new setup required.
Sensible billing
Top up via QRIS, bank transfer, or e-wallets. Official invoices for businesses.
Production-ready
Rate limiting, monitoring, automatic failover, and built-in circuit breaker.
API compatibility
Drop-in for the SDK you already use.
Just change baseURL to Genfity. All methods, parameters, and response formats are identical.
Pricing
3 ways to pay, pick what fits.
Credits for exploration, Subscription for production, or PAYG for dynamic traffic.
7 Days Unlimited Token + Opus 4.7/4.8 PRO
7 Days
- 5,000 requests / period
- 1,000 requests/day (RPD)
- Unlimited daily credit quota
- Unlimited period credit quota
- 120 requests/min
- 6 concurrent
- 600,000,000 tokens / period
- Unlimited Token Input & Output
1 Days Claude + All Models
1 Days Unlimited Token Claude Model Lite
7 Days Unlimited Token Opus 4.8/4.7 GPT 5.5
7 Days China Models
30 Days Developer Lite
30 Days
- 10,000 requests / period
- 500 requests/day (RPD)
- 3,000 credits/day
- 30,000 credits / period
- 120 requests/min
- 6 concurrent
- 1,000,000,000 tokens / period
- Unlimited Token Input & Output
30 Days Developer Pro
30 Days
- 10,000 requests / period
- 500 requests/day (RPD)
- 4,500 credits/day
- 45,000 credits / period
- 120 requests/min
- 6 concurrent
- 1,000,000,000 tokens / period
- Unlimited Token Input & Output
Supported Models
Models currently published in the gateway.
| Model | Context | Capability | Copy |
|---|---|---|---|
Auto genfity/auto | 200,000 | ToolsVisionStream | |
Claude Haiku 4.5 genfity/claude-haiku-4.5 | 200,000 | ToolsVisionStream | |
Claude Opus 4.6 genfity/claude-opus-4.6 | 1,000,000 | ToolsVisionStream | |
Claude Opus 4.7 genfity/claude-opus-4.7 | 1,000,000 | ToolsVisionStream | |
Claude Opus 4.8 genfity/claude-opus-4.8 | 1,000,000 | ToolsVisionStream | |
Claude Sonnet 4.6 genfity/claude-sonnet-4.6 | 1,000,000 | ToolsVisionStream | |
Deepseek 3.2 genfity/deepseek-3.2 | 128,000 | ToolsStream | |
Deepseek v4 Flash genfity/deepseek-v4-flash | 1,000,000 | ToolsStream | |
Deepseek v4 Pro genfity/deepseek-v4-pro | 1,000,000 | ToolsStream | |
GLM 5 genfity/glm-5 | 128,000 | ToolsStream | |
GLM 5.1 genfity/glm-5.1 | 202,752 | ToolsStream | |
GPT 5.3 Codex genfity/gpt-5.3-codex | 400,000 | ToolsVisionStream | |
GPT 5.4 genfity/gpt-5.4 | 400,000 | ToolsVisionStream | |
GPT 5.4 Mini genfity/gpt-5.4-mini | 400,000 | ToolsVisionStream | |
GPT 5.5 genfity/gpt-5.5 | 400,000 | ToolsVisionStream | |
Gemini 3.1 Pro genfity/gemini-3.1-pro | 1,000,000 | ToolsVisionStream | |
KIMI k2.5 genfity/kimi-k2.5 | 262,144 | ToolsVisionStream | |
KIMI k2.6 genfity/kimi-k2.6 | 262,144 | ToolsVisionStream | |
MiniMax M3 genfity/minimax-m3 | 1,000,000 | ToolsVisionStream | |
Minimax M2.5 genfity/minimax-m2.5 | 1,000,000 | ToolsVisionStream | |
Qwen3.6 Flash genfity/qwen3.6-27b | 262,144 | ToolsVisionStream | |
Qwen3.6 Plus genfity/qwen3.6-plus | 1,000,000 | ToolsVisionStream | |
Subagent genfity/subagent | 200,000 | ToolsVisionStream | |
Xiaomi Mimo v2.5 genfity/mimo-v2.5 | 1,000,000 | ToolsVisionStream | |
Xiaomi Mimo v2.5 Pro genfity/mimo-v2.5-pro | 1,000,000 | ToolsVisionStream |
Features
Production-ready, out of the box.
Everything you need to run AI in production—without building it yourself.
Rate Limit per API Key
RPM and TPM per key, isolated per environment.
Real-time Analytics
Live dashboard: cost, tokens, latency per model.
Automatic Failover
Automatic fallback if a provider has downtime.
Streaming Support
SSE streaming for all chat models.
Tools / Function Calling
Compatible with OpenAI/Anthropic tool use spec.
Indonesian Support
WIB support team via WhatsApp & email.
FAQ
Frequently asked questions.
Key details before using AI Gateway for your product.
Can it really be cheaper?
Yes. Use affordable models for common requests, reserve premium models for tasks that need strong reasoning. Save 40-60% on mixed use cases.
Can I keep using popular SDKs?
100% compatible with OpenAI SDK. Just change baseURL to api.genfity.com/v1—all methods and responses are identical.
How does billing work?
3 models: Credit (package top-up), Subscription (unlimited monthly), or PAYG (USD balance per token). Pick what fits your usage.
Is it production ready?
Yes. Rate limits, failover, monitoring, and circuit breaker are built-in. 99.9% uptime target with formal SLA for Pro/Enterprise plans.
Start with one API key. Scale to many models when ready.
Sign up free with starter credit. No USD credit card needed.
