01//
Multi-provider
OpenAI, Anthropic, Google models behind one API key. Failover between providers built-in.
ONE API KEY · 8 LEADING LLM MODELS
The single platform to buy, track, and route LLM tokens
Route requests to GLM 5.2, DeepSeek V4, Kimi and GPT-OSS with automatic provider failover and real-time token tracking.
Trusted by
/ CURSOR/ CLINE/ AIDER/ CONTINUE/ ZED/ CODY/ ROO CODE
// Why Jetnight
Production-ready infrastructure for top LLM models.
Simplify your LLM stack. Route requests dynamically, track token usage in real time, and gain absolute resilience.
02//
Token tracking
Real-time usage stats, per-key limits, detailed logs. Spot leaks instantly.
03//
Drop-in compatible
Works with Claude Code, Cline, Kilo Code, Opencode and any OpenAI SDK — no code changes.
04//
Smart Failover
Automatic fallback to backup models or providers if latency spikes or rate limits occur.
05//
Edge Caching
Global edge caching reduces latency for repeated prompts to under 50ms.
// How it works
From signup to first request in 60 seconds.
01
Buy tokens
Pick a bundle and pay with SBP, card, or crypto. Your API key lands instantly.
02
Point your SDK
Set the base URL to api.jet-night.com and use any OpenAI-compatible client.
03
Route & ship
Switch models with one line. Failover, limits, and logs run automatically.
// Models
Every model, one endpoint.
Route to the best LLM for the job — switch models with one line, no new SDK, no extra keys.
Custom...
Moonshot...
OpenAI...
Zhipu AI...
Google...
MiniMax...
Anthropic...
DeepSeek...
ModelContextSpeedFeaturesStatus
More models added weekly. All routed through a single OpenAI-compatible endpoint.
// Pricing
Buy token bundles
Buy once — tokens stay on your account until you spend them. No subscriptions.
99₽
Mini
55Mtokens
Starter pack: flash/mini models for trying the API and light work.
Features:
- Quick start
- 1 API key
- Base models
- 20 RPM
Speed20 RPM
ModelsGemini Flash, GPT-mini, Haiku, Kimi K2, DS Flash
249₽-20%
Start
170Mtokens
Work pack: Sonnet, mid GPT-5.x, Grok and light reasoning.
Features:
- Chat + code
- 3 API keys
- 40 RPM
- Request history
Speed40 RPM
ModelsMini + GPT-5.2/5.4, Sonnet 4.6, Grok-4, DS R1, o4-mini
599₽-40%
Pro
550Mtokens
Hero plan: full frontier (5.6 / Opus 5 / Fable 5) without Enterprise price.
Features:
- Frontier: 5.6 Luna / Sol / Terra
- Opus 5 + Fable 5
- 10 API keys · 100 RPM
- Reasoning + tools + vision
- Best value
Speed100 RPM
ModelsGPT-5.6 Sol/Luna/Terra · Opus 5 · Fable 5 · GPT-5.5 · Grok 4.5 · Kimi K3
1 790₽-80%
Max
5Btokens
Full catalog + huge token pool for teams and heavy pipelines.
Features:
- Everything in Pro
- 25 API keys
- 300 RPM
- Full catalog
- Team access
Speed300 RPM
ModelsFull catalog (107 models)
4 990₽SLA
Enterprise
∞ / 30 days
Unlimited usage, SLA and dedicated terms for business.
Features:
- Unlimited tokens 30 days
- Custom routing
- Unlimited RPM
- Audit + SLA
- Priority support
SpeedUnlimited
ModelsFull catalog + unlimited
access
Models by plan
Model access depends on your plan. Pro unlocks frontier (5.6, Opus 5, Fable 5). Max and Enterprise get the full catalog.
| Model group | Mini99₽ | Start249₽ | Pro599₽ | Max1 790₽ | Ent.4 990₽ |
|---|---|---|---|---|---|
| Gemini Flash, GPT-mini, Haiku, Kimi K2, DS Flash | |||||
| GPT-5.2 / 5.4, Sonnet 4.6, Grok-4, DS R1, o4-mini | |||||
| GPT-5.5, Grok 4.5, Kimi K3, DS V4 Pro | |||||
| GPT-5.6 Sol / Luna / Terra | |||||
| Claude Opus 5 + Fable 5 | |||||
| Full catalog — 107 models |
Out-of-plan model returns HTTP 403. Authenticated /v1/models is filtered by plan. Free matches Mini.
yesno
Custom token amount
Choose how much you need. More = cheaper per 1M.
tokens
1M1.2B2.5B3.7B5B
// Order details
Amount50M
Volume discount-6% discount
Rate per 1M1.7₽ / 1M
Total price85₽
Checkout now// Drop-in
Drop it into any OpenAI-compatible client.
Point your existing SDK at the Jetnight Router endpoint and keep shipping. One key, 8 models, zero refactor.
1Create account and get your API key
2Drop the code into any OpenAI-compatible client
3Start building with 8 AI models
ENDPOINTrouter.jet-night.com/api/v1
AUTH HEADERBearer jn_sk_...
FORMATOpenAI Compatible
// Code example
1234567891011
import OpenAI from "openai";const client = new OpenAI({ apiKey: "jn_sk_YOUR_KEY", baseURL: "https://router.jet-night.com/api/v1"});const response = await client.chat.completions.create({ model: "jetnight-fast", messages: [{ role: "user", content: "Build something fast." }]});main#27997D
SBPFaster Payments System
secondary#D7D7D8
RU cardsMir · Visa · Mastercard
popular#26A17B
CryptoUSDT · BTC · ETH
telegram#007AFF
CryptobotPayments via Telegram
stars#A855F7
xRocketTelegram Stars
delivery#F59E0B
InstantKey in your account right after payment
// Payments
Pay your way. Key delivered instantly.
SBP, Russian cards, crypto, Cryptobot and xRocket — pick what works and your API key lands in your account the moment payment clears.
All transactions are processed securely. API license keys are generated in real-time and added directly to your dashboard as soon as the payment confirmation is received.
Answers to the questions that come up most.
Learn how Jetnight works, what models are supported, how to set up integration, and what to expect as the platform grows.
What is Jetnight?
Jetnight is a unified API gateway for LLMs. One key, one endpoint, one bill — with failover, usage tracking, and instant delivery.
Which models are supported?
GLM 5.2, Claude Opus 4.8, DeepSeek V4, Kimi K2.7, MiniMax M3, and more added weekly. All through the same OpenAI-compatible endpoint.
Do you store my prompts or data?
No. Jetnight acts as a stateless proxy. Your prompts, completions, and keys are never logged or stored on our servers.
Is there a free trial or demo credits?
Yes. Every new account receives free starting credits to test the API and explore all available models instantly.
Can I use Jetnight in commercial applications?
Absolutely. Jetnight is built for production loads, offering high uptime, load balancing, and low latency for businesses.