8 models · rates as of 2026-09-25
Pick a model. Keep your key.
Every alias works in Claude Code, Codex, the API and the chat app. Switch any time with /model — your key, balance and history stay the same.
Claude Opus 5.5
Anthropic's Opus flagship. Claude Code default.
- Input
- $5.00
- Output
- $25.00
- Context
- 1M
GPT-6 Astra
OpenAI's frontier model for the most demanding work.
- Input
- $12.50
- Output
- $62.50
- Context
- 1M
GPT-6 Sol
Fast, capable all-rounder. Chat default.
- Input
- $2.50
- Output
- $12.50
- Context
- 1.1M
GPT-6 Luna
The budget model for quick, high-volume work.
- Input
- $0.13
- Output
- $0.63
- Context
- 1.1M
GPT-4o
The reliable multimodal classic.
- Input
- $3.13
- Output
- $12.50
- Context
- 128K
DeepSeek V4 Pro
DeepSeek's flagship. Strong coding and reasoning at a low price.
- Input
- $2.18
- Output
- $4.35
- Context
- 1M
DeepSeek V4.1 Flash
Fast, very cheap DeepSeek model for everyday coding.
- Input
- $0.55
- Output
- $2.74
- Context
- 1M
Gemini 3.8 Flash
Google's fast thinking model, straight from Google.
- Input
- $0.94
- Output
- $4.69
- Context
- 1M
Auto
Tools or long context go to Claude Opus; short chat goes to the cheapest healthy GPT-6. Billed at the model it picks.
At a glance
Side by side
| Model | Best for | In /1M | Out /1M | Context | Typical turn | Eval |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 aethene/claude-opus | Claude Code sessions, refactors, debugging across a codebase. | $5.00 | $25.00 | 1M | $0.068 | pending |
| GPT-6 Astra aethene/gpt-6-astra | Hard planning, deep reviews and a strong second opinion. | $12.50 | $62.50 | 1M | $0.197 | pending |
| GPT-6 Sol aethene/gpt-6-sol | Chat, Codex sessions, planning and everyday coding. | $2.50 | $12.50 | 1.1M | $0.039 | pending |
| GPT-6 Luna aethene/gpt-6-luna | Quick questions, summaries, boilerplate and high-volume work. | $0.13 | $0.63 | 1.1M | $0.002 | pending |
| GPT-4o aethene/gpt-4o | Existing prompts and apps already tuned for GPT-4o. | $3.13 | $12.50 | 128K | $0.098 | pending |
| DeepSeek V4 Pro aethene/deepseek-v4-pro | Everyday coding and long-context work when cost matters. | $2.18 | $4.35 | 1M | $0.023 | pending |
| DeepSeek V4.1 Flash aethene/deepseek-v4-flash | Quick edits, summaries, boilerplate and high-volume work. | $0.55 | $2.74 | 1M | $0.009 | pending |
| Gemini 3.8 Flash aethene/gemini-3-8-flash | Long documents, images, and fast reasoning at a low price. | $0.94 | $4.69 | 1M | $0.014 | pending |
Our bar is at least 95% on a tool-call eval (multi-file edit, test-run-fix loop, long cached prompt, streaming cancel, sub-agent spawn), re-run weekly. Models are in beta until their scores are measured and published here.
Auto
aethene/auto picks for you
Requests with tools or long context go to Claude Opus; short chat goes to whichever of GPT-6 Sol and Luna is healthy and cheaper. You're billed at the model it picked, and your request log shows which one that was.