{"agent":"skeptic","tool":"read_page","args":{"url":"https://darwincompute.com/"},"result":"<untrusted_content source=\"https://darwincompute.com/\">\nTitle: Darwin Compute\n\ndarwin\nHow it picks\nGPUs\nSetup\nDocs\nConnect wallet\nOpen app\ndarwin\n\nKeep working when your model's limit runs low.\n\nOne API key for every model. Darwin keeps your best model for hard work, moves easy requests to cheaper ones as its limit runs down, and runs its own GPU model when everything else is spent.\n\nOpen the app\n↗\nRead the docs\n$DARWIN\n5VnbrK…Wnpump\npump.fun\nYour best model keeps the hard work\n\nDarwin grades every request before it goes out and sends it down your list of models: the hardest to the top, the easy ones to cheaper models as the top one's limit runs low. A conversation stays on one model unless that model runs out.\n\nYour best model\n\nPlans, hard bugs, long refactors. Darwin keeps these here until its limit is nearly gone.\n\nClaude Fable 5.1, GPT-6 Astra\n\nA mid-size model\n\nMost coding work and tool calls once the top model's budget is past half.\n\nClaude Sonnet 5.5, GPT-6.1 Sol, Gemini 3.1 Pro\n\nA small, fast model\n\nRenames, summaries, lookups and other easy requests.\n\nClaude Haiku 4.5, Gemini 3.8 Flash, DeepSeek V4.1 Flash\n\nDarwin's GPU\n\nQwen3 30B on Darwin's own hardware. No key needed, and it catches everything when the rest are out.\n\nPaid with burned $DARWIN\n\nA model of its own\n\nDarwin runs Qwen3 30B on its own GPUs. It answers with no provider key at all, and when every model on your list is out, the work goes there instead of stopping. Every reply names the model that wrote it.\n\nAccess is paid for by burning $DARWIN from your wallet. Each burn turns into credit for the API and the MCP server, and credit buys upgrades: Pro moves an agent to gpt-oss-120b with a 128K context, and warm hours keep a GPU running so it never waits for a cold start.\n\nSetup is one setting\n\nConnect a wallet or make a key in the app, then point the base URL in Claude Code, the OpenAI or Anthropic SDK, or any client that takes one at Darwin. Add your own provider keys whenever you want Darwin to route between your models.\n\nRead the setup guide\nWhich models does it work with?\n+\nHow does it decide how hard a request is?\n+\nHow do I pay for Darwin?\n+\nWhich limits does it watch?\n+\nDoes a conversation stay on one model?\n+\nWhat happens to my keys?\n+\nWhy not just lower the effort setting?\n+\ndarwin\n\nPrices from OpenRouter's public model list, checked 3 October 2026. Dashboard counters by rareui.com.\n\nHow it picks\nGPUs\nSetup\nDocs\nDashboard\nX\n\nLinks:\nhttps://darwincompute.com/#top\nhttps://darwincompute.com/#how\nhttps://darwincompute.com/#compute\nhttps://darwincompute.com/#setup\nhttps://darwincompute.com/docs\nhttps://x.com/darwincompute\nhttps://darwincompute.com/app?connect=1\nhttps://darwincompute.com/app\nhttps://pump.fun/coin/5VnbrKp28Qs9CAH6PyZdxBNvLxZcbeX3YBsozgWnpump\nhttps://rareui.com/\n</untrusted_content>","isError":false,"images":[]}