{"slug": "show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits", "title": "Show HN: Virtual Private LLM, fixed fee with no usage or token limits", "summary": "Solheim AI launched Virtual Private LLM (VPL), a flat-fee service providing private, EU-hosted LLM instances with no token or usage limits, starting at €15.00 per month for one instance with a 64k context window. The service runs open-weight models like Qwen3.6-35B-A3B and DeepSeek V4 Flash on EU-owned infrastructure, with an OpenAI-compatible API for integration with tools like Cline and VS Code.", "body_md": "Issue 01 — a VPS for language models · sign-up open\n\n# Rent the machine.\n\nNot the tokens.\n\nA private LLM instance on EU hardware, for a flat monthly fee. Nothing counts down and nothing resets — the only limits are how many requests you run at the same moment and how much context each one gets, and you set both yourself.\n\nNo token meter.\n\nNo rolling usage window.\n\nNo request cut off halfway\n\nthrough your task.\n\nNo surprise bills — your instance\n\nat one fixed monthly cost.\n\nEU hardware, EU jurisdiction.\n\n01 / The idea\n\n## If you've bought a server, you know this product.\n\nOne project gets one VPL: one endpoint, one key, two numbers to size it.\n\nA slider sets how many instances the project runs, and an instance is one request in flight. Three instances means three things at once: your agent, your editor, your test loop. A second dial sets the context window each of those requests gets. The API is OpenAI-compatible, so Cline, ZooCode, VS Code BYOK and your own backend all talk to it unmodified.\n\n-\n01\nOn a VPS — vCPU + RAM\n\nInstance count\n\nOne slider, one dial. Inside that size, run as much as you like, all day.\n\n-\n02\nOn a VPS — no request quota\n\nNo usage window\n\nNothing counts down and nothing resets. You are never locked out mid-task.\n\n-\n03\nOn a VPS — flat monthly invoice\n\nFlat monthly invoice\n\nNo surprise bills: a busy sprint costs exactly what a quiet one costs. Tokens are counted for you and never billed to you.\n\n-\n04\nOn a VPS — OS image\n\nOpen-weight model\n\nPick and pin the model. Swap it by changing one string, no migration.\n\n-\n05\nOn a VPS — one server, one host\n\nOne project, one VPL\n\nKeys, usage and limits stay per project — nothing is shared by accident.\n\n-\n06\nOn a VPS — datacenter region\n\nEU-only region\n\nNothing leaves EU jurisdiction — GDPR and AI Act posture by construction.\n\n02 / Migration\n\n## Two minutes, two fields.\n\nNothing to install, nothing to rewrite.\n\n-\n01\n### Create a project\n\nA project is one VPL: its own endpoint, its own key, its own usage. Name it after the thing it serves.\n\n-\n02\n### Set the instance count\n\nOne slider decides how many requests run at once. Move it up for a week of parallel agents, back down after. That is the whole capacity model.\n\n-\n03\n### Point your tool at it\n\nBase URL and key into Cline, ZooCode, VS Code BYOK, or any OpenAI SDK. Nothing else in your setup changes.\n\n```\nexport OPENAI_BASE_URL=https://api.solheim.ai/v1\nexport OPENAI_API_KEY=slh_live_9f3c…\n\n# beyond your instance count, calls queue.\n# they do not 429 and they do not cost extra.\n```\n\n03 / Sovereignty\n\n## EU-hosted, all the way down.\n\nSovereignty is an architecture, not a checkbox.\n\n-\n01\nInfrastructure\n\nGPUs at EU-owned providers in EU datacenters. No US hyperscaler in the serving path, so no CLOUD Act exposure to write into your DPA.\n\n-\n02\nModels\n\nOpen weights under permissive licenses, pinned per instance and run by us. Your prompts are not training data — not ours, not anyone's.\n\n-\n03\nRouting\n\nRequests never leave the region, and this page loads nothing from a third-party origin — no US CDN, no analytics, no fonts phoning home.\n\nThe library\n\n-\n01\nQwen3.6-35B-A3B\n\nqwen3.6-35b-a3b\n\nThe default for day-to-day work: diffs, tests, refactors and tool-calling agents. A mixture-of-experts model: 35B parameters in total, three billion of them active per token, price-value like nothing else.\n\n-\n02\nDeepSeek V4 Flash\n\ndeepseek-v4-flash-0731\n\nFor more demanding work: near frontier open-weight performance on your own reserved compute\n\n-\n+\nRequest a model\n\nopen weights · permissive licence\n\n04 / Pricing\n\n## Two dials: how many, how much context.\n\nInstances × context window × a flat monthly rate. Still no tokens, no tiers and no overage — there is nothing on this invoice that can surprise you.\n\n- Your window\n- 64k\n- Model\n- one per project\n- Beyond your count\n- queues, never 429\n- Overage\n- there isn't any\n\n€15.00 /month · 1 × 64k × €15.00\n\nExcludes VAT, which is added at checkout based on where you are.\n\n[Start at this size](#get-started)\n\nBeyond ten, or beyond 128k\n\nDedicated & enterprise\n\nA dedicated GPU or a MIG slice of one — no shared neighbours, no contention with anyone else's agents. A 256k window lives here too: it is four times the cache of the largest shared size, so it gets its own card rather than a queue. Private networking, a signed DPA and invoicing rather than a card. Priced against the hardware you need, not per request.\n\nLet's talk\n\n[Contact us](mailto:hello@solheim.ai?subject=Dedicated%20capacity)\n\n05 / Readership\n\n## Who it's for.\n\nSame product, same dials, same arithmetic.\n\n-\n01\nDevelopers running coding agents\n\nYou live in Cline, ZooCode or VS Code and you hit the window most weeks. Three instances is an agent, an editor and a test loop running at once — for as long as the day lasts.\n\n-\n02\nFounders who need an inference backend\n\nYour product needs a private LLM behind it, with a bill that doesn't move when traffic does. Point your backend at the same endpoint and size it the same way.\n\nIt is one product, not two. What changes between these is what you point at the endpoint — not the pricing, not the API, not the machine.\n\n06 / From the blog\n\n## Notes\n\n[All notes](/blog)\n\n[The Break-Even Point Between Metered Tokens and Your Own LLM Per-token pricing is the right way to buy inference until a certain volume. This is the arithmetic for finding that volume, and what to do on either side of it. Read](/blog/breakeven-metered-tokens-vs-private-llm)\n\n[How to Replace Claude Code With Your Own Private LLM Swap the closed-source agent for an open harness and a model endpoint you control. Same workflow, no usage windows, no data leaving your control. Read](/blog/replace-claude-code-self-hosted-llm)", "url": "https://wpnews.pro/news/show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits", "canonical_source": "https://solheim.ai", "published_at": "2026-08-13 16:04:36+00:00", "updated_at": "2026-08-13 16:13:42.370912+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-infrastructure", "ai-policy"], "entities": ["Solheim AI", "Qwen3.6-35B-A3B", "DeepSeek V4 Flash", "Cline", "ZooCode", "VS Code"], "alternates": {"html": "https://wpnews.pro/news/show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits", "markdown": "https://wpnews.pro/news/show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits.md", "text": "https://wpnews.pro/news/show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits.txt", "jsonld": "https://wpnews.pro/news/show-hn-virtual-private-llm-fixed-fee-with-no-usage-or-token-limits.jsonld"}}