Liquid LFM2.5 8B
Liquid's efficient 8B Liquid Foundation Model tuned for long-context reasoning on-device.
LFM2.5 8B is an open-weight text model built by Liquid. It takes a text prompt and replies with generated text. It runs entirely on your own hardware through Ollama: no API key and no per-use fees.
- Released
- May 2026
- Pricing
- Free, runs on your hardware
Samples
Generated with LFM2.5 8B via Ollama. The same three prompts run against every text model in the catalog, shown verbatim, so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API is a set of rules that lets one piece of software send requests and receive responses from another without needing to see how the other works; it acts as an interface, allowing different programs to communicate securely and efficiently.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
The startup should build its own user authentication system rather than rely on a hosted provider; however, this approach is dangerously costly because it demands extensive security expertise and ongoing maintenance that would otherwise be redirected toward product innovation.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
Executive leadership must recognize that predicting the company’s AI spending is difficult because several forces create uncertainty. First, rapid technological change means new tools and capabilities appear frequently, causing cost patterns to shift unexpectedly. Second, data quality remains variable; without reliable information the effectiveness of any model can differ widely, leading to unpredictable expenses. Third, evolving regulations and ethical expectations impose compliance costs that cannot be forecast in advance. Fourth, recruiting and retaining skilled talent introduces budget lines that fluctuate with market demand. Finally, integrating AI with existing legacy systems often uncovers hidden expenditures that were not evident during planning. To address this, leaders should adopt a flexible budgeting approach that funds projects incrementally based on clear outcomes rather than fixed long‑term commitments. They should also create governance structures that define decision authority, risk limits, and performance indicators for each initiative. Encouraging collaboration among business units, legal counsel, and technology teams will align spending with strategic priorities while providing early signals when assumptions fail. By applying these steps, the organization can retain control over its AI investments despite the inherent volatility of the field. These actions will help executives make decisions, reduce financial risk, and ensure that AI initiatives deliver measurable value to the company.
Liquid
Liquid develops Liquid Foundation Models (LFM) — a new architecture optimized for efficient long-context reasoning.
www.liquid.ai ↗Liquid AI's mixture-of-experts reasoning model with 24B total and 2.45B active parameters. It always reasons before answering and is built for agents that call tools.
Liquid AI's successor to D1, a structured decision model served as a System One endpoint for choice, score and yes/no answers about text and images.
Liquid AI's largest decision model, served as a System One endpoint: it reads text and images and answers typed questions with a choice, a score or a yes/no probability.
Liquid's 24B foundation model with efficient long-context reasoning.
Liquid AI's first omni-modal LFM3 model (2.2B parameters), built for on-device use; the app sends it text and images.
Liquid's structured decision model, served as a System One endpoint for choice, score and yes/no answers.