Aethene
Get started
ModelsPricingCalculatorSetupStatusSign in

Aethene for developers

Every top model. One key.

Claude Opus 5.5, GPT-6, DeepSeek V4 and Gemini 3.8 in Claude Code, Codex and a web chat — one key, one balance, ten ready-made skills. Set up in three minutes.

Get started See pricing
$ curl -fsSL https://3-229-227-242.sslip.io/install.sh | sh
Claude Code Codex Web chat Pay per token
Refactor the billing module and add tests
Done — 3 files changed and all 24 tests pass. Here is what ran:
Refactored billing/settle.go
aethene/claude-opus · claude-code · 1.9s
$0.068
Titled the session
aethene/gpt-6-luna · side request · 0.4s
$0.0007
Reviewed the diff
aethene/gpt-6-sol · second opinion · 2.4s
$0.039
Ask any model… Opus 5.5
Works with Claude Code Codex CLI Anthropic SDK OpenAI SDK Azure AI Foundry Google Vertex AI

Set up in three minutes

One command installs the CLI. It backs up your settings, points Claude Code at Aethene and adds ten skills.

Terminal

## INSTALL

curl -fsSL aethene.com/install.sh | sh

## SIGN IN

aethene login

## CONNECT

aethene install claude

Every model in /model

Switch models inside Claude Code without touching a config file. Your key, balance and history stay the same.

aethene/auto
Claude Opus 5.5$5.00/M
GPT-6 Astra$12.50/M
GPT-6 Sol$2.50/M
GPT-6 Luna$0.13/M

Chat in the browser

The same models and balance in a web chat, with the cost of every reply right under it.

ChatLive
Summarise this PR for the changelog
Adds per-key spend caps and a CSV export for request logs.

$0.0098 · aethene/gpt-6-sol

See every request

Tokens, cost, latency and route for each call. Never the prompt.

“Fix the flaky retry test”$0.47

Skills built in

Ten skills for Claude Code and the chat app. Type / and go.

/api-client-generator/bug-triage/commit-messages/docs-writer

Caps and failover

Per-key spend caps you control. If a model is busy, its fallback answers before any output reaches you.

Can this CI key spend without limit?
No — “ci-runner” stops at its $20 monthly cap. Revoke it any time.
No prompts stored

Room for the whole repo

Up to 1.1M tokens of context, and prompt caching so long sessions stay cheap.

GPT-6 Sol
1.1M context · cache reads $0.25/M

Models

Every model, priced to the token.

Supplier list price plus a fixed 25%, published per model. Rates as of 2026-09-25. Switch any time with /model — your key and balance stay the same.

Claude Opus 5.5

Anthropic's Opus flagship. Claude Code default.

Input
$5.00
Output
$25.00
Context
1M
aethene/claude-opus≈ $0.068 / turn

GPT-6 Astra

OpenAI's frontier model for the most demanding work.

Input
$12.50
Output
$62.50
Context
1M
aethene/gpt-6-astra≈ $0.197 / turn

GPT-6 Sol

Fast, capable all-rounder. Chat default.

Input
$2.50
Output
$12.50
Context
1.1M
aethene/gpt-6-sol≈ $0.039 / turn

GPT-6 Luna

The budget model for quick, high-volume work.

Input
$0.13
Output
$0.63
Context
1.1M
aethene/gpt-6-luna≈ $0.002 / turn

GPT-4o

The reliable multimodal classic.

Input
$3.13
Output
$12.50
Context
128K
aethene/gpt-4o≈ $0.098 / turn

DeepSeek V4 Pro

DeepSeek's flagship. Strong coding and reasoning at a low price.

Input
$2.18
Output
$4.35
Context
1M
aethene/deepseek-v4-pro≈ $0.023 / turn

DeepSeek V4.1 Flash

Fast, very cheap DeepSeek model for everyday coding.

Input
$0.55
Output
$2.74
Context
1M
aethene/deepseek-v4-flash≈ $0.009 / turn

Gemini 3.8 Flash

Google's fast thinking model, straight from Google.

Input
$0.94
Output
$4.69
Context
1M
aethene/gemini-3-8-flash≈ $0.014 / turn

Auto

Tools or long context go to Claude Opus; short chat goes to the cheapest healthy GPT-6. Billed at the model it picks.

aethene/autoCompare all →

How it works

Your tools. Our gateway. Every model.

Aethene sits between your agent and the models. It checks your key and balance, streams every token straight through, and records what it cost.

Your tools Claude Code Codex CLI Web chat Your code
Aethene gateway Key, balance and caps checked Tokens streamed, never buffered Fallback model before the first byte Cost recorded per request
Models Claude Opus 5.5 GPT-6 Astra GPT-6 Sol GPT-6 Luna GPT-4o DeepSeek V4 Pro DeepSeek V4.1 Flash Gemini 3.8 Flash

API

Three formats. One key.

Anthropic Messages, OpenAI Chat Completions and the Responses API — with streaming and tool use — so existing SDKs work by changing the base URL.

Anthropic Messages
curl https://api.3-229-227-242.sslip.io/v1/messages \
  -H "x-api-key: $AETHENE_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{"model": "aethene/claude-opus",
       "max_tokens": 1024,
       "messages": [{"role": "user",
         "content": "Review this diff"}]}'
OpenAI Chat Completions
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.3-229-227-242.sslip.io/v1",
    api_key=os.environ["AETHENE_KEY"],
)
client.chat.completions.create(
    model="aethene/gpt-6-sol",
    messages=[{"role": "user",
               "content": "Plan the migration"}],
)
Responses API (Codex)
# ~/.codex/config.toml — written for you by
#   aethene install codex
model = "aethene/gpt-6-sol"
model_provider = "aethene"

[model_providers.aethene]
base_url = "https://api.3-229-227-242.sslip.io/v1"
wire_api = "responses"
0prompts or responses stored by the gateway90 daysof request metadata, then deleted25%over supplier list price, published per model$5minimum top-up; unused paid credit refundable for 14 days

Pricing

Estimate your cost
Starter
$20/ 30 days
Excluding tax · cancel any time
Get Starter

Includes:

  • 3,000 units every 30 days
  • ≈ 272 typical Claude Code turns
  • Every model in Claude Code, Codex and chat
  • Wallet rates after units run out
Pro
$50/ 30 days
Excluding tax · cancel any time
Get Pro

Everything in Starter, with:

  • 7,500 units every 30 days
  • ≈ 681 typical Claude Code turns
  • Pay by card or from your balance
  • Units reset every 30 days
Power
$135/ 30 days
Excluding tax · cancel any time
Get Power

Everything in Pro, with:

  • 20,000 units every 30 days
  • ≈ 1,818 typical Claude Code turns
  • Sized for heavy daily use
  • Cancel any time; units last to period end

Prefer to pay as you go? Top up from $5 and pay per token at published rates.

Start with the wallet

FAQs

How is Aethene different from a Claude subscription?
It's pay-as-you-go access to several top models on one key — in Claude Code, Codex, your own code and a web chat. Heavy all-day Claude users are usually better off with a Claude Max subscription; Aethene suits light and medium use, teams and anyone mixing models.
Which tools does it work with?
Claude Code and Codex CLI out of the box (aethene install claude / codex), plus any client that speaks the Anthropic Messages, OpenAI Chat Completions or Responses API.
Where do I chat?
In the web chat at app.aethene.com — same account, same balance, with every reply's cost shown.
How much does it cost?
Top up from $5 and pay per token at published rates, or pick a plan from $20 per 30 days. A typical Claude Code turn costs about $0.068.
Can I control spending?
Yes. Every key can have a monthly cap, an expiry date and a list of allowed models, and you can revoke it instantly. Turn on auto top-up if you never want a pause.
How do you handle my data?
The gateway never stores prompts or responses — only metadata like token counts and cost, kept for 90 days. Chat history is stored for the chat app until you delete it.

One key. Three minutes.

$ curl -fsSL https://3-229-227-242.sslip.io/install.sh | sh
Get started